The Sun is "spotty and impure." At least, that's how the disappointed Italian astronomer Galileo Galilei described it when he turned his telescope on our local star early in the 17th century. Galileo was disillusioned to discover that the Sun was covered in bizarre dark blotches that were seen to migrate across the solar disk. Many scientists at that time argued that these were satellites of the Sun transiting its face. But Galileo and another astronomer, David Fabricius, came up with an alternative interpretation. The spots, they concluded, were on the Sun's surface and their migration was evidence of its rotation.
Galileo and Fabricius were right. Sunspots, as these afflictions were soon called, are indeed features on the photosphere --the solar surface. But for some 300 years, nobody knew much more than that. It wasn't until the advent of 20th-century spectroscopy that astronomers gleaned their first real clue as to the nature of sunspots. The key, it turned out, was magnetism.
The Sun is a seething magnetic cauldron because its outermost 30 percent (ending just below the photosphere) is a turbulent convective zone. Gas is heated at the bottom of this layer and rises in cells to the surface, visible on the Sun as granulation, to transfer that heat into space. When the gas cools, it sinks into the interior, where it is reheated and the cycle starts again. The magnetic field is a side effect of all this. When coupled with the Sun's rotation, electrical currents created as the ionized gas or plasma rises and falls generate a magnetic field. However, the resultant field is not uniform because the Sun is not a solid body--its equator spins faster than its poles. This differential rotation has the effect of stretching the magnetic field such that over time, north-south field lines become drawn out in an east-west direction. Hot plasma, twisting as it rises, regenerates the north-south lines but also braids the magnetic field into rope-like bundles with buckles and kinks. Where the kinks are greatest the magnetic field can be up to 1,000 times more intense than average. These fields inhibit the flow of gases and make the local surface about 1,500[degrees] Celsius (2,700[degrees] Fahrenheit) cooler than the surrounding regions. From a distance, these cool areas show up as dark patches--the sunspots.
If the Sun has spots, the natural question to ask is, what about other stars? From a theoretical point of view the answer is clear. Cool stars--those with convective zones--probably have spots if they rotate fast enough to generate magnetism. However, hot stars, lacking convective motion, are expected to be spotless. Of course the stars are too distant to allow us to directly test this theory. But despite the great gulf that separates us from our stellar neighbors, astronomers now know for certain that some of them do indeed harbor dark starspots similar to sunspots. The trick is to identify tracers of magnetic activity on the Sun and then to look for them on other stars too. The obvious place to look is in solar and stellar spectra.
Spots: The Telltale Signs
A stellar spectrum tells us what a star is made of. As the star's light passes through the gases in its atmosphere, various atoms and molecules absorb different portions of that starlight. A particular element will absorb light of only certain wavelengths, resulting in the appearance of dark absorption lines in
the spectrum. Similarly, under certain conditions some elements emit rather than absorb light at particular wavelengths and produce bright emission lines. Thus, a star's spectrum holds information from which astronomers can identify the composition and other properties of the star's atmosphere.
How does this help in the hunt for starspots? Well, spectroscopy of the Sun has shown that regions of enhanced magnetic activity (sunspots) emit strongly at wavelengths that correspond to the element calcium, the so-called H and K lines (S&T: December 1996, page 40). So if stellar spots in general are related to magnetic activity as they are on the Sun--and there is every reason to believe this to be the case--the presence of spots on other stars can be deduced merely by looking for this calcium H and K fingerprint. Astronomers started applying this technique decades ago when they began a monitoring program with the Mount Wilson 60-inch telescope. It turns out that many stars exhibit the calcium-emission signature. They are magnetic and will probably have spots. But there is an added surprise. Repeated observations of some of these stars have shown not only that they are magnetic but also that the strength of their calcium emission and therefore their magnetic fields vary periodically on time scales of several years--something also seen in the Sun. The number of sunspots, active regions, and flares present on our star at any given time varies with a period of about 11 years--the sunspot cycle. In other words, some stars appear to have magnetic cycles just as the Sun does. By analogy with the Sun, this strengthens the case for starspots.
Another way starspots can be inferred is from stellar rotation. If an active star has spots on it, its light level will change subtly as it rotates. For example, when a relatively spotty hemisphere is in view, the star appears slightly dimmer than when its clear side is facing us. In fact, there exist stars whose light levels change so greatly as they spin that they must be covered in truly enormous blemishes--not just spots but vast continents of darkness.
For many decades, these and similar indirect methods were the only tools available for starspot studies. Not anymore. Since 1982, astronomers have been perfecting a technique that enables them not only to infer the presence of spots but also to actually image them and map their distribution across a given stellar surface--to see them, to a certain extent. It's a technique known as Doppler tomography and, surprisingly, its roots lie within the medical profession.
Stellar CAT Scans
Most people have heard of computer-aided tomography--the CAT scans that doctors use to produce cross-sectional images of organs inside the body. The scanning machine beams X-rays through the body and a detector records an image as the rays emerge. Then the machine makes a small rotation in a plane that slices through the organ under study, and another image is taken from the new position. A full rotation of the machine results in a series of images of the specimen taken from a range of angles covering a full 360 degrees. The final step is a computer algorithm that analyzes and merges the data to produce a tomogram --a diagnostic picture showing a slice through the organ.
Obviously rotation is one of the key elements in tomography. But rather than rotating the detector, one could just as easily build up a tomogram by turning the patient. In 1982, this is exactly what two astronomers (Steven S. Vogt and G. Donald Penrod, Lick Observatory) set out to do. Their "patient" was not a person, though--it was a star, already naturally revolving in space. Their technique, which has become greatly modified and is now widely used, relies on the Doppler shifts that absorption lines suffer as a star spins. These lines move toward the blue part of the spectrum when the absorbing region approaches the observer and are redshifted when the region is in recession. Thus the star's motion modulates the shapes and positions of its spectral lines. With some very clever mathematics, the changes in the spectrum as a function of time--in essence, changes in the viewing angle--can be used to reconstruct an image of the star's surface.
Vogt and Penrod first used this method to assemble an image of the surface of a star in an RS Canum Venaticorum binary system. These systems frequently comprise two G-class Sun-like stars closely orbiting each other with periods ranging from hours to weeks. Most RS CVn stars spin much more quickly than the Sun, which takes almost a month to complete a rotation. Such rapid rotation ensures that these stars' convective zones generate substantial magnetic fields. Vogt's and Penrod's remarkable images of the subgiant component (2.83-day rotation) in the system HR 1099 showed several large spots, some covering up to 10 percent of the surface. One major starspot straddled the pole while others wandered near the equator. Immediately, two differences emerge when we compare these stellar spots with sunspots. The first is the scale--HR 1099's spots are 100 to 200 times more extensive than the Sun's. The second is that, while one of HR 1099's patches straddles a pole, sunspots by contrast are never seen outside solar latitudes of 40[degrees].
Since Doppler tomography was first developed, astronomers have imaged many stars with this technique and certain patterns are emerging. It is now clear that magnetic activity varies in direct relation to the spin period. The faster the rotation, the greater the magnetic field, and the larger the spots. Indeed some stars have spots that blanket up to 50 percent of their surface. Polar spots are also seen to be common, especially in the fastest rotators.
But it isn't just RS CVn stars that have substantial magnetic fields. The BY Draconis stars, usually G- or M-class dwarfs, exhibit photometric and spectral variations related to extensive magnetic activity. Some flare stars also show spots. T Tauri stars--young, rapidly spinning stellar objects still contracting toward the main sequence and thus not yet undergoing the hydrogen burning that drives mature stars--are also very magnetically active.
Until fairly recently, tomography could be used to image only rapidly rotating stars--as long as they didn't spin too rapidly. Because the method requires long exposure times to generate exceptionally good-quality data with high signal-to-noise ratios, the surface features of very rapid rotators become blurred. On the other hand, tomography was initially not sensitive enough to resolve the smaller spots found on slow rotators. But techniques are improving all the time. The latest craze for extracting an image from tomographic data is something called least-squares deconvolution. It has been used extensively by Andrew Collier Cameron (University of St. Andrews, Scotland), and Jean Francois Donati (Observatoire Midi-Pyrenees, France). With this method, Collier Cameron and Donati have been able to resolve spots that extend for only 2[degrees] or 3[degrees] across a stellar surface--similar in size to extremely large sunspots. And least-squares deconvolution can even detect changes in the pattern and distribution of those spots caused by differential rotation. To date, astronomers have captured and analyzed the differential rotation on three stars: AB Doradus, PZ Telescopium, and RX J1508.6-4423. No doubt more will follow in due course.
Theory expects that cool, rapidly rotating stars will have spots. Photometric and spectroscopic observations support that theory. And now tomography has revealed actual images of the spots themselves, proving that these features are not unique to Galileo's "spotty and impure" Sun. They are a result of stellar magnetism and, as such, may be present on many, if not most, Sun-like stars.
MARK A. GARLICK is a freelance writer, illustrator, and former astronomer based in Brighton on England's south coast. He is currently writing and illustrating his first book, Story of the Solar System. You can see more of his paintings and digital artwork online at www.space-art.co.uk.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||sunspots on other stars|
|Author:||Garlick, Mark A.|
|Publication:||Sky & Telescope|
|Date:||Mar 1, 2001|
|Previous Article:||Today's science of the sun.|
|Next Article:||The little rose of winter.|