All astronomical observations are subject to unwanted "noise:" signals that can mask the output from the object to be studied. Noise can come from the telescope and detector, from the atmosphere, from other objects in the sky, or from intrinsic variations in the object to be studied. The PI will use the technique known as Gaussian Processing to model and reduce the interference of astronomical noise to facilitate the study of planets around other stars. These planets are studied by their "eclipses," or transits, in front of the star. Many stars, however, vary in their light output, and this can cause a misinterpretation of the transit data. The method to be developed will help reduce the noise due to the variability of the star. A graduate student will be trained, and undergraduate students will be involved in the research. The computational techniques developed will be applicable to fields beyond astronomy.
The proposer will develop a method for finding small transiting exoplanets and/or exomoons in the presence of stellar variability. Stellar brightness variations are correlated across time and wavelength, and the fast computational method he proposes to develop will be employed to use these correlations to distinguish between transits and random dips in stellar brightness. This technique will be extended to account for outliers and can be applied to other types of exoplanet and astronomical data, as well as data in other fields. The code written under this proposal will be made available to the community through open-access development.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
