Astronomy › Section L · Analysing starlight and the stellar census › Topic L.7
Diameters of stars
Through almost any telescope a star is only a point of light, yet astronomers know that some stars are smaller than Earth and others would swallow the orbit of Mars. Eclipses, interferometers and the laws of radiation reveal sizes no ordinary image can show.
Points of light
The Sun is the one star whose disc is easy to see and measure. Almost every other star is so far away that even the largest single telescopes show it only as a point; the small blur in an image comes from turbulence in Earth's atmosphere, not from the star's real disc. Yet radius is a key property, a clue to how a star shines and what stage of life it has reached. This topic gives four ways to find it.
Angular size and true size
An object's angular diameter \(\theta\) is the angle it covers on the sky. For small angles, the true diameter is
with \(\theta\) in radians and \(d\) the distance. The Sun's diameter is \(2R_\odot \approx 1.39 \times 10^9\) m (one solar radius, \(R_\odot\), is about \(6.96 \times 10^8\) m). At 1 AU, \(1.496 \times 10^{11}\) m, that gives \(\theta = 0.0093\) radians, or about 0.53°, which matches what we see. The Sun is about 109 times Earth's diameter. The largest angular diameters of other stars are only about 0.05 arcseconds (an arcsecond is 1/3600 of a degree), far too small for a normal telescope image.
Stars hidden by the Moon
As the Moon moves around Earth, its sharp edge sometimes passes in front of a star, an event called a lunar occultation. A point source would vanish instantly, but a star with a real disc fades over a small fraction of a second as the edge crosses it. Because the Moon's motion across the sky is known precisely, fast light measurements of that fade give the star's angular diameter, and with the distance its true size. The method is precise but limited: it works only for stars that lie near the Moon's path, close to the ecliptic, such as Aldebaran, Antares, Regulus and Spica.
Interferometry
An interferometer combines light from two or more separated telescopes so that they act, for measuring fine detail, like one telescope as wide as their separation. In 1920 Albert Michelson and Francis Pease used one mounted on the 100-inch telescope at Mount Wilson to measure the angular diameter of Betelgeuse, about 0.047 arcseconds, the first measured for any star beyond the Sun. Modern arrays such as CHARA in California and the Very Large Telescope Interferometer in Chile, with telescopes up to hundreds of metres apart, have measured the diameters of hundreds of stars and even imaged the surfaces of a few, such as Betelgeuse and Altair.
Eclipsing binaries
If a binary's orbit is seen almost edge-on, each star passes in front of the other once per orbit, and the combined light dips. A graph of brightness against time is called a light curve. The classic example is Algol in Perseus, which fades noticeably every 2.87 days for about ten hours. In 1783 John Goodricke suggested that a dimmer companion was eclipsing it; in 1889 Hermann Vogel's spectra confirmed that Algol is a binary. The deeper eclipse happens when the hotter star, which gives more light per square metre, is hidden. Only a small fraction of binaries happen to be lined up to eclipse as seen from Earth.
Diameters from eclipse timing
Take an edge-on orbit in which a small star passes behind a large one. First contact is when the small star's edge touches the large star's edge and the light begins to drop; at second contact the small star is completely hidden; at third contact it starts to reappear; at fourth contact the eclipse ends. Between first and second contact the small star moves its own diameter, and between first and third contact it moves the large star's diameter. The Doppler shifts of both stars give their relative speed \(v\), so
Worked example: timing an eclipse
In an edge-on eclipsing binary the stars' relative speed is 200 km/s. The total eclipse begins 1.50 hours after first contact, and third contact comes 12.0 hours after first contact. 1. Convert times: 1.50 h = 5400 s and 12.0 h = 43,200 s. 2. Small star: \(D = 200 \times 5400 = 1.08 \times 10^6\) km, a radius of \(5.4 \times 10^5\) km, or 0.78 solar radii. 3. Large star: \(D = 200 \times 43{,}200 = 8.64 \times 10^6\) km, a radius of 6.21 solar radii. 4. Check: the large star comes out bigger, and the light should take as long to recover after third contact as it took to fall, 1.50 hours.
Radius from luminosity and temperature
The Stefan–Boltzmann law says each square metre of a star's surface emits \(\sigma T^4\) watts, with \(\sigma = 5.67 \times 10^{-8}\) W m⁻² K⁻⁴. Multiply by the surface area \(4\pi R^2\) to get the luminosity:
Comparing with the Sun (about 5,770 K) removes the constants:
So two stars of equal temperature but 100 times different luminosity differ in radius by a factor of 10. The method needs only a luminosity, which requires the distance, and a temperature, from colour or spectrum, so it works for many more stars than eclipses do.
Worked example: a supergiant and a white dwarf
Find the radii of a red supergiant with \(L = 100{,}000\ L_\odot\) and \(T = 3500\) K, and a white dwarf with \(L = 0.01\ L_\odot\) and \(T = 20{,}000\) K. 1. Supergiant: \(\sqrt{100{,}000} = 316\) and \((5770/3500)^2 = 2.72\), so \(R \approx 860\ R_\odot\), about 4.0 AU. 2. White dwarf: \(\sqrt{0.01} = 0.1\) and \((5770/20{,}000)^2 = 0.0832\), so \(R \approx 0.0083\ R_\odot\), about 5,800 km, roughly Earth's size. 3. Check: the cool star is luminous only because it is vast, and the hot star is dim only because it is tiny.
The range of stellar sizes
Measurements show that most stars are within a factor of a few of the Sun's size, but the extremes are astonishing. White dwarfs are about as big as Earth, around a hundredth of the Sun's radius. Red dwarfs range from about a tenth to two-thirds of the Sun's radius, and brown dwarfs are close to Jupiter's size. Giants are roughly 10 to 100 times the Sun's radius, and red supergiants reach hundreds to over a thousand. Recent estimates for Betelgeuse are around 700 to 1,000 solar radii: placed at the Sun, its surface would lie beyond the orbit of Mars, at 1.52 AU, though inside Jupiter's, at 5.2 AU.
Common mistakes to avoid
First, taking the size of a star's image as its true size: the blur is caused by the atmosphere and the telescope. Second, using degrees instead of radians in \(D = \theta d\). Third, mixing up eclipse intervals: first to second contact gives the small star's diameter, first to third gives the large star's. Fourth, forgetting the square and fourth power in \(L = 4\pi R^2 \sigma T^4\): doubling the temperature at fixed radius multiplies the luminosity by 16. Fifth, assuming a luminous star must be hot, when a cool star can be luminous by being enormous.
Key takeaways
Apart from the Sun, stars appear as points, so diameters need special methods. Lunar occultations time how quickly a star fades behind the Moon's edge. Interferometers, first used on Betelgeuse in 1920, measure tiny angular diameters directly. Eclipsing binaries give diameters from contact times and orbital speed: D = v × time. The Stefan–Boltzmann law, L = 4πR²σT⁴, gives a radius from luminosity and temperature. Stars range from Earth-sized white dwarfs to red supergiants hundreds or more than a thousand times the Sun's radius.
Flashcards
Flashcard
The angle a star's disc covers on the sky
Test your understanding
Question 1 of 5
Two stars have the same surface temperature, but one is 400 times as luminous as the other. How do their radii compare?
Question 2 of 5
A star has twice the Sun's radius and twice the Sun's surface temperature. How luminous is it compared with the Sun?
Question 3 of 5
In an edge-on eclipsing binary the relative orbital speed is 100 km/s, and first to second contact takes 2.0 hours. What is the smaller star's diameter?
Question 4 of 5
Why can lunar occultations measure the diameters of only some stars?
Question 5 of 5
A star has a surface temperature of 20,000 K but only 1% of the Sun's luminosity. What must be true of it?
