Zwicky's Stellar Sleuths
Stars have an array of different sizes, colors, and temperatures. They also rotate, blow away their outer layers in stellar winds, and in some cases have completely different processes happening in their cores. Despite these many differences, if you go outside on a dark night and look up at the sky, the vast majority of stars will look nearly identical. The distinct properties of individual stars can be captured by a telescope. All we can measure from these stars is the light they emit (at least this was the case for hundreds and hundreds of years, a new type of telescope can detect gravitational waves from some types of stars; checkout Gravity Spy for more information).
Everything we know about stars and other celestial bodies comes from their light. Zwicky's Stellar Sleuths examines how the brightness of stars changes as a function of time in order to classify the stars that we observe. Once the type of star is known, we can infer other information about the star such as its distance, age, or, in some cases, how it formed, or, how it will die.
We astronomers are a creative bunch, which is why we call a star that gets brighter or fainter a "variable star". Variability can happen for a multitude of reasons – some stars follow a specific set pattern, while others change somewhat randomly and without much warning. Variability patterns can repeat on timescales of a few minutes, or a star can lay dormant for thousands and thousands of years before erupting in a violent and energetic flash. We need your help classifying these stars so we can better understand how stars change throughout their lives. We discuss some examples below, but if you aren't interested in those details, go forth and classify!
We are in the (second) golden age for the study of celestial objects that change on human timescales. The recent proliferation of wide-field survey telescopes have led to a rapid expansion in the number of known variable stars over the past decade. These efforts have allowed us to find faint events, fast events, and small-amplitude events at rates that were previously unimaginable. A lot of this work will culminate in a few years with the start of the Legacy Survey of Space and Time conducted by the Vera C. Rubin Observatory, which is expected to find more than 20 million variable stars on its own.
Zwicky's Stellar Sleuths aims to find and classify variable stars that have been identified by the Zwicky Transient Facility (ZTF). The observations taken by ZTF are capable of finding a vast array of variable stars, such as cataclysmic variables, Cepheids, RR Lyrae, and eclipsing binaries. Each of these systems is provides useful clues for studying phenomena ranging from the shape and size of stars to galaxy formation to measuring the expansion rate of the Universe. ZTF is also sensitive to rare objects, such as short-period white dwarf binaries, large-amplitude, radial-mode hot subdwarf pulsators, and ultracompact hot subdwarf binaries.
To highlight one source class, we are discovering and characterizing the population of so-called ultra-compact binaries (UCBs), which have two stellar-mass compact objects with orbital periods less than 1 hour. Many of the UCBs emit gravitational waves in the milliHertz regime with sufficient strain for the upcoming Laser Interferometer Space Antenna (LISA) to detect. These LISA “verification sources” will serve as crucial steady-state sources of gravitational waves that will not only verify that LISA is operating as expected, but also themselves serve as probes of binary stellar evolution, white dwarf structure, Galactic structure, accretion physics, and general relativity.
There are SO MANY stars in the sky – identifying the most interesting ones is really challenging. ZTF has already observed more than 3 billion sources, and we need your help to classify them! Your efforts will help us learn more about the astrophysics of stellar variations, and very likely identify new types of variability as well!
At the moment, variable star classification efforts with computers, using machine learning algorithms and other types of artificial intelligence, are somewhat limited. While we would like to use these tools to classify the billions of stars that have been observed, we currently lack a large set of variables that has been correctly classified – an essential step in building the machine learning algorithms. With your help, we can construct the large data sets that the computers need to "learn" how to tell apart different types of variables. As the algorithms improve their ability to identify the different types of stars, we will provide you the variables that are most difficult to classify. With humans and machine working back and forth in this manner, we will together discover entirely new classes of astronomical objects.