Astro Working Group III: Time Series Analysis for Exoplanets & Gravitational Waves: Beyond Stationary Gaussian Processes
Group Leaders: Ben Farr (Astro, U. Chicago, chair), Eric Feigelson (Astro, PSU), G. Jogesh Babu (Stat, PSU), Rebekah Dawson (Astro, UC Berkeley); Soumen Lahiri (Stat, NCSU).
Description: The characterization of time-domain data is a fundamental component of astronomy. We will explore new techniques for modeling such data, both evenly and unevenly sampled in time, with the goal of enabling new discoveries in exoplanet and gravitational wave astronomy.
Big Questions:
- Gravitational wave detectors are typically assumed to have stationary and Gaussian noise. However, the noise spectral density changes on long timescales and there are frequent short duration noise artifacts (i.e., "glitches") that can mimic gravitational wave signals. How can we build more robust noise models and exploit information in auxiliary environmental channels?
- Ground-based photometric surveys, with billions of irregular light curves, detect stellar variability from a multitude of sources, from planetary transits, to supernovae. How do we effectively separate these classes in irregularly sampled data?
- How can we maximize the sensitivity of space-based photometric surveys for detecting small planets in the presence of stellar activity?
- How can we measure planet masses robustly despite observational data with outliers (due to stellar activity affecting either Doppler data or transit times)?