Citizen ASAS-SN

Frequently Asked Questions

---

Here we try to answer some of the most-commonly asked questions about Citizen ASAS-SN and the science behind it.

---

• WHAT IS ASAS-SN?
  The All-Sky Automated Survey for Supernovae (ASAS-SN) is the first and only project that images the entire visible sky on a nightly basis. Each ASAS-SN unit has four 14 cm aperture telescopes with cooled, back-illuminated CCD cameras equipped with g-band filters. ASAS-SN presently consists of five mounts, hosted by Las Cumbres Observatory, with one unit at each of MacDonald Observatory (Texas), Haleakala (Hawaii), the South African Astronomical Observatory, and two at Cerro Tololo International Observatory (Chile). A sixth mount is being commissioned in western China.

• WHAT IS A LIGHT CURVE?
  Simply put, a light curve is a plot of a star's brightness over time. ASAS-SN observes the entire visible sky every night using telescopes in both the northern and southern hemispheres. We use these observations to make brightness measurements of about 100 million stars throughout the duration of our project.

• WHAT WAVELENGTHS OF LIGHT ARE SEEN BY ASAS-SN?

  ASAS-SN currently monitors the entire night sky in the optical wavelength range. Fun fact: Humans also observe their world in the optical! To make these observations scientifically useful, we limit the photons that we capture by using a filter. ASAS-SN uses Sloan g-band filters with an effective central wavelength of 480.3 nm, and a FWHM of 140.9 nm.

• HOW DOES ASAS-SN OBSERVE AND PROCESS ITS DATA?
  To accomodate the characteristics of the ASAS-SN telescopes, we divide the sky into numerous fields, with each field being observed by at least one ASAS-SN camera every night. Some fields are observed by more than one camera. As of now, each field has been observed about 2000 times.
  We process our observations through a procedure known as Image Subtraction.

  • Reference images are built up for each ASAS-SN field after averaging observations made over a long period of time. These reference images contain information on the average brightness of the stars in that field.

  • When a field is observed every night, we make three dithered 90 sec exposures, with a typical depth of g = 18 mag. We co-add these exposures to increase the signal-to-noise ratio of our data.

  • We subtract the reference image from the new observations to obtain a subtracted image. This image provides a wealth of variability information about stars, including the identification of transient objects that were previously not seen in the reference images. A perfectly constant star will not exist in the subtracted image following the image subtraction procedure. However, transient objects like supernovae will show up in the subtracted images. ASAS-SN uses the subtracted images to identify transients in the nearby universe. We also use the subtracted images to make brightness measurements of stars using a photon counting procedure known as photometry.

  Image subtraction yields excellent results, particularly towards very crowded regions of the Galaxy and enables us to obtain excellent light curves of variable stars in the Milky Way.

• HOW FAR AWAY ARE THE STARS OBSERVED BY ASAS-SN?

  ASAS-SN can observe stars that are as close as a few hundred light years and the most distant stars are in small satellite galaxies orbiting the milky way, particularly the Magellanic Clouds, which are over 100,000 light years away. Stars that are closer than a few hundred light years are very bright and are saturated in our data, so we don't have reliable light curves of these sources. On the contrary, stars that are very far away get fainter and are progressively harder to observe given that we use small 14 cm telescopes.

• ASAS-SN ALREADY CLASSIFIED VARIABLE STARS IN THE V-BAND DATA. WHY ARE YOU DOING THIS AGAIN WITH THE g-BAND?

  Yes, using machine learning methods we have already classified variable stars in our archival V-band data (2013-2017) obtained with just 2 ASAS-SN units. These classifications are publicly available on our Variable Stars Database.

  However, with the g-band data, we have made several improvements over our initial V-band data set.

  • ASAS-SN g-band data is obtained using 5 units (vs. 2 for the V-band), with the latitude spread of the five mounts helping to greatly suppress diurnal and annual aliasing of the light curves, which improves the identification of variable stars.
  • In the g-band data, we can look at fainter objects (g~18 mag, ~100 million stars) compared to the V-band data (V~17 mag, ~62 million stars)
  • We have improved our rate of data collection from ~2-3 days in the V-band to ~20 hours in the g-band, allowing us to study and discover more variable stars in greater detail than we did before.

• WHAT HAPPENS TO THE CLASSIFICATIONS I PROVIDE?

  They're stored with those provided by everyone who takes part. The Citizen ASAS-SN team will carefully analyze your work to provide aggregated classifications of variable stars. All results will eventually be made public for anyone to use.

• WHY DO YOU NEED SO MANY PEOPLE TO HELP?

We have a lot of light curves! While machine learning is good at some things, it remains bad at a number of things that people are just really, really good at. In particular, people are really good at picking out outliers. Some of these outliers are a result of our automated pipeline doing something wrong, and some of them will be genuine astrophysical outliers. Other outliers will be rare combinations of known variables, like a pulsating star in a eclipsing binary. These rare objects can be very useful scientific toolsand allow us to pursue new science.