It’s 2 a.m., the witching hour. I write this from the Dark Sky Observatory at Appalachian State, where I am on the first night of a four-night observing run with colleagues from Georgia and Kentucky. One of their students is with them, and I have a soon-to-be astronomy major visiting from Newport, N.C., as well, hoping to get a taste of research.
Normally nowadays, I would have the telescope observing automatically as I sleep at home, having set up things remotely in the early evening. The sequence of images would be programmed, with timers set to turn off the telescope at dawn, and an electronic finger held to the air to close the dome if bad weather moves in. Another tireless electronic eye would be staring at the sky all night, waiting to close the dome at the light of dawn.
But tonight, it’s old-school observing, necessitated by the targets and changing cloudiness. We are taking images of binary stars – stars that orbit each other. Our targets take about half a day to do this – imagine, two suns that complete a lap in 12 hours.
We happen to be edge-on to their orbit so even though all of these binaries are too far away to appear as anything more than an unresolved, single dot of light, each star blocks the view of the other star every orbit. Studying the changing brightness of these eclipsing binaries yields their component stars’ sizes, masses (“weights”) and temperatures. It’s about the only good source of information on the nature of other suns.
We get the data by taking a series of digital images through different color filters and measure the star brightness on each image. A graph of the brightness versus time – a so-called “light curve” – can be modeled with a computer program to determine the two stars’ physical properties. By observing in different colors, we get the temperatures, since, in the world of physics, bluer is hotter.
If clouds edge in, we have to increase the exposure times. We may abandon one binary system and move to another if our target sets too low in the sky. We monitor the brightness with quick, real-time measurements to see when the eclipses are occurring.
We may decide to do calibration images as dawn sets in. Our stellar data images include the effects of dust on the optics that cast shadows on the images and variations in sensitivity across the imaging detector. We can remove these effects by imaging a uniformly bright surface such as the dawn sky.
But it will be a struggle then to do that imaging: It’s now 3 a.m. It is difficult now to write this. Let’s hope for some clouds about dawn.
Daniel B. Caton is a physics and astronomy professor, and director of observatories at Appalachian State University. Email: email@example.com. More on this month's column: www.upintheair.info.