Yale Researchers Develop Innovative Data-Driven Method to Detect Exoplanets

Yale University researchers have developed a new, data-driven approach to detecting exoplanets. Credit: Illustration by Michael Helfenbein/Yale University

Researchers from Yale University have discovered a data-driven technique to find out distant planets and to fine-tune the search for worlds identical to Earth.

The innovative technique has been described in a study reported in The Astronomical Journal on 20 December and is based on mathematical methods originating from physics research. Instead of attempting to filter out the signal “noise” from stars around which exoplanets orbit, Yale researchers analyzed all the signal information jointly to perceive the intricacies within its structure.

It requires nothing but the data itself, which is a game changer. Moreover, it allows us to compare our findings with other, traditional approaches and improve whatever modeling assumptions they use.

John Wettlaufer, A.M. Bateman Professor of Geophysics, Mathematics and Physics at Yale University

The hunt for exoplanets (planets outside the solar system) has drastically increased in recent years. The attempts have been motivated, to some extent, by a passion to detect Earth analogs that could also have the ability to support life.

Researchers have used various methods in their attempt, such as direct imaging, pulsar timing, and measuring the speed at which galaxies and stars move either away from or toward Earth. However, these methods, either in combination or individually, pose difficulties.

Primarily, these difficulties can be overcome by eliminating extraneous data - noise - that does not match existing models explaining the way in which planets are anticipated to behave. In this conventional interpretation of noise, searches can be hindered by data that conceal or imitate the exoplanets.

Wettlaufer and his collaborators decided to look for exoplanets in the same manner in which they had sorted through satellite data to detect complex alterations in Arctic sea ice. The method has been formally named as “multi-fractal temporally weighted detrended fluctuation analysis” (MF-TWDFA). It analyzes data over all time scales and extracts the underlying procedures associated with them.

A key idea is that events closer in time are more likely to be similar than those farther away in time. In the case of exoplanets, it is the fluctuations in a star’s spectral intensity that we are dealing with.

John Wettlaufer, A.M. Bateman Professor of Geophysics, Mathematics and Physics at Yale University

Benoit B. Mandelbrot and Katepalli Sreenivasan pioneered the application of multi-fractals in science and mathematics at Yale. For performing a highly proficient search for exoplanets, the team sought the advice of Yale astrophysicist Debra Fischer, who has been a pioneer of various methods applied in the field.

The accuracy of the methodology was validated by the researchers by testing it against simulation data and observations of a known planet orbiting a star in the constellation Vulpecula, which is located nearly 63 light years from Earth.

The first author is Sahil Agarwal, graduate student in the Yale Program in Applied Mathematics. The second author is Fabio Del Sordo, joint postdoctoral fellow at Yale and in Stockholm.

The research was funded by grants from NASA, the Swedish Research Council, as well as a Royal Society Wolfson Research Merit Award.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback