Ask an astronomer what the hardest thing to do is in astronomy, and chances are they will say ‘measuring distance accurately.’ It is surprisingly difficult to take the light from stars we see and match them to a correct distance. In the past we have used several different methods depending on how close a star is to us. For the nearest stars we use parallax, which looks at the change in a star’s position as the Earth is on opposite sides of it’s orbit.
All other methods rely on what we call the standard candle approach. Let’s say you had a light bulb that you move far away from me. I can measure its brightness, but it doesn’t tell me it’s distance. However, if you tell me the bulb’s intrinsic brightness, ie that the bulb is a 40-watt light bulb, I can use some mathematics to figure out it’s distance. This is because there is a direct relationship between apparent brightness, intrinsic brightness, and distance. In astronomy, the hard part is figuring out the intrinsic brightness, and we generally do this with computer models of stars based on their temperature and composition. But this does produce a lot of uncertainty, since the models are approximations at best.
Astronomers at Cambridge University have developed a new method to determine distances with incredible precision, without requiring any stellar models at all. Their idea is to look at pairs of stars that have exactly the same spectra. With so many stars in the galaxy, this is a surprisingly common scenario. If one of the stars in the pair are close enough to use the parallax method and find its distance, we can also measure its apparent brightness and combine the two to find the intrinsic brightness of the star. We can then look at the more distant star, measure its apparent brightness, use the same intrinsic brightness (since they are twins), and calculate its distance.
Using a population of 600 stars that have high-resolution spectra, the astronomers identified 175 pairs of twin stars. For each twin, one of the stars was close enough for a parallax measurement.
There is a catch however. Stellar spectra are very busy and difficult to compare. A single stellar spectrum can contain as many as 280,000 data points. for this reason, the researchers chose 400 spectral lines to make their comparisons. The lines they chose give the most distinguishing information about the star, like looking at two photographs of crowds of people, and using people that stand out to determine if they are the same shot.
The next task for the researchers is to find a list of all stars for which distance data is available, and then compare their spectra to all stars with inaccurate distance data, looking for twins to help fill in the gaps. Repeating this process will help them slowly build up a complete distance catalogue for stars in the Milky Way, allowing us to better understand the structure and evolution of our home galaxy.