Exoplanet Detection with Bayesian Blocks

Abstract:

Discovery of exoplanets is important for understanding our place in the universe. One fruitful method for locating exoplanets is by the measurement of the drop in flux of the planet’s host star when the planet passes between the observer and the star – also known as transiting. The light-curves produced by planets transiting in front of their parent star can help us determine how our own solar system compares to others and the general behavior of stellar systems by giving us an understanding of the components of a typical stellar system. Now that the time-tagged flux data from thousands of these events is publicly searchable from the Kepler mission on the Mikulski Archive for Space Telescopes, it is easier than ever to conduct meaningful analysis on these transit events through the use of the IDL programming language. In more detail, I used IDL to develop an automated algorithm that finds a transit event in the data recorded by the Kepler Mission. This is carried out by a Bayesian algorithm that determines where the transit occurs by searching for a statistically significant change in the flux of the light-curve. The final product is a single program module that will give a measurement of the transit start and end times, the transit durations, and period of the planet’s orbit, with uncertainties.