Astronomy › Section M · Celestial distances › Topic M.2
Surveying the stars: parallax
Parallax is the only way to measure a star's distance with pure geometry, using Earth's orbit as a surveyor's baseline. Every other method for stellar and galactic distances is calibrated, directly or indirectly, against it.
From the planets to the stars
Radar fixed the solar system's scale, as Fundamental units of distance explained, but radar echoes from stars are out of the question. Voyager 1, about 173 astronomical units (AU) away in late 2026, has barely started the journey: the nearest star is roughly 269,000 AU off. To reach the stars, astronomers borrow the surveyor's method of triangulation. A surveyor who cannot cross a river to a distant tree sights the tree from two points a known distance apart, called the baseline. The tree's direction differs slightly from the two points, and that apparent shift, caused purely by the change of viewpoint, is called parallax.
A shift that shrinks with distance
Hold a finger at arm's length and look at it with one eye, then the other. It jumps against the background. Bring the finger closer and the jump grows; move it farther away and the jump shrinks. Your eyes, a few centimetres apart, are the baseline. The rule is general: for a fixed baseline, the farther the object, the smaller the parallax angle. The only way to measure very small angles reliably is to make the baseline as long as possible. Ptolemy used Earth's rotation to triangulate the Moon, and later astronomers used baselines thousands of kilometres apart on Earth for the nearer planets. Stars need far more.
Earth's orbit as the baseline
Over six months Earth moves from one side of its orbit to the other, a baseline of 2 AU, about 300 million km. A nearby star seen in January and again in July appears to move slightly against much more distant background stars, tracing a tiny ellipse each year. Astronomers define the parallax angle \(p\) as half of that total shift, which is the angle the radius of Earth's orbit, 1 AU, makes as seen from the star. Even for the nearest stars, \(p\) is less than one arcsecond, where 1 degree = 60 arcminutes = 3600 arcseconds (written 3600″). A one-arcsecond angle is roughly the width of a small coin seen from about 4 km.
Why the ancients saw nothing
Aristotle argued that if Earth moved around the Sun, the stars should show parallax, and since none was seen, Earth must be still. Tycho Brahe made the same argument in the late 1500s after measuring star positions with the naked eye to about one arcminute. The reasoning was valid but the premise was wrong: stars are so far away that even the largest shift is about 80 times smaller than anything Tycho could detect. Success came only with precise telescopes. In 1838 Friedrich Bessel announced the parallax of 61 Cygni, and Thomas Henderson (Alpha Centauri) and Friedrich Struve (Vega) reported parallaxes at about the same time.
The parsec
For small angles, the distance is the baseline divided by the angle in radians: \(d = \frac{1\ \mathrm{AU}}{p}\). One arcsecond is \(\frac{\pi}{180 \times 3600} = 4.848 \times 10^{-6}\) radians, so a star with a parallax of exactly 1″ lies at \(\frac{1}{4.848 \times 10^{-6}} = 206{,}265\) AU. That distance is the parsec (pc), short for parallax-second, the unit first met in Numbers in astronomy and light travel time: about \(3.09 \times 10^{16}\) m, or 3.26 light-years (ly). The payoff is a beautifully simple rule:
A parallax of 0.1″ means 10 pc. The rule works only with \(p\) in arcseconds and \(d\) in parsecs.
Worked example: the nearest star
Proxima Centauri has a measured parallax of 0.768″. How far away is it, in parsecs, light-years and AU? 1. Distance in parsecs: \(d = \frac{1}{0.768} = 1.30\) pc. 2. In light-years: \(1.30 \times 3.26 = 4.25\) ly. 3. In AU: \(1.30 \times 206{,}265 = 2.69 \times 10^5\) AU. 4. Check: the parallax is a little less than 1″, so the distance should be a little more than 1 pc, and it is. No star is closer than 1 pc, so every stellar parallax is smaller than 1″. Proxima is a faint red dwarf, the most common kind of star; the bright pair Alpha Centauri A and B lies slightly farther, at about 4.37 ly.
Worked example: a parallax and its uncertainty
A space telescope measures a star's parallax as 5.0 milliarcseconds (mas), with an uncertainty of 0.5 mas. Find the distance and its range. 1. Convert: 1 mas = 0.001″, so \(p = 0.0050\)″. 2. Distance: \(d = \frac{1}{0.0050} = 200\) pc, about 652 ly. 3. Range: \(p\) could be 0.0055″ or 0.0045″, giving \(\frac{1}{0.0055} = 182\) pc and \(\frac{1}{0.0045} = 222\) pc. 4. Interpret: a 10% uncertainty in parallax gives roughly a 10% uncertainty in distance. Because \(d = 1/p\), a small angle error matters more and more as \(p\) shrinks, which is why parallax works best for nearby stars.
Leaving the atmosphere
From the ground, turbulent air smears each star into a shimmering blob, and for over a century the best parallaxes were good to about 0.01″, enough for reliable distances out to only about 20 pc. Space solved this. The European Space Agency's Hipparcos satellite, launched in 1989 and operating until 1993, measured about 118,000 stars to roughly 0.001″ (one milliarcsecond), giving good distances out to a few hundred light-years. Its name honours the Greek astronomer Hipparchus and also stands for High Precision Parallax Collecting Satellite.
Gaia: a billion-star survey
ESA launched Gaia in December 2013 to a station about 1.5 million km beyond Earth, where it scanned the whole sky repeatedly. It ended science observations on 15 January 2025 and was switched off on 27 March 2025, but its data are still being released. Gaia Data Release 3 (June 2022) listed about 1.8 billion sources, about 1.47 billion with parallaxes, precise to a few hundredths of a milliarcsecond for bright stars. As of November 2026, Data Release 4, based on 66 months of observations, is scheduled for 2 December 2026, with a final release expected no earlier than the end of 2030. Gaia gives useful parallaxes across a large part of the Milky Way.
Common mistakes to avoid
First, using the total six-month shift instead of half of it: the parallax is defined with a 1 AU baseline. Second, plugging milliarcseconds straight into \(d = 1/p\); convert to arcseconds first. Third, thinking a larger parallax means a more distant star; it is the reverse. Fourth, confusing parallax with proper motion: parallax is a yearly back-and-forth wobble from Earth's motion, while proper motion is the star's own steady drift across the sky. Fifth, forgetting that 1 pc is 3.26 ly, not 1 ly.
Key takeaways
Parallax is the apparent shift of a nearby star as Earth orbits the Sun; the parallax angle p is half the six-month shift. Distance follows from geometry alone: d (pc) = 1/p (arcsec). One parsec is 206,265 AU, about 3.09 × 10¹⁶ m or 3.26 ly. The first stellar parallaxes came in 1838. All parallaxes are under 1″; the nearest star, Proxima Centauri, has p = 0.768″ and lies 1.30 pc away. Space missions avoid atmospheric blurring: Hipparcos reached about 0.001″, and Gaia, retired in 2025, has measured about 1.5 billion parallaxes, with its fourth data release due in December 2026.
Flashcards
Flashcard
Apparent shift in an object's direction caused by a change of viewpoint
Test your understanding
Question 1 of 5
A star has a parallax of 0.0400 arcseconds. How far away is it?
Question 2 of 5
Between January and July a nearby star shifts by a total of 0.40″ against the background stars. What is its distance?
Question 3 of 5
Star A has twice the parallax of star B. What can you conclude?
Question 4 of 5
Why do space missions such as Hipparcos and Gaia measure parallaxes far better than ground telescopes?
Question 5 of 5
Which statement about Gaia is correct as of November 2026?
