Automated Detection and Removal of Hidden Systematic Noise in Kepler/K2 Photometry with Principle Component Analysis

We present a novel method for detrending systematic noise from time series data using a combination of Principal Component Analysis (PCA) and Fast Fourier Transforms (FFT). This method is demonstrated on time series data obtained from the inaugural campaign of the Kepler/K2 mission, as well as three objects of interest from Campaign 4. This method could improve measurements of stellar flare statistics by distinguishing low-energy flares from surrounding noise. We test the detrending on periodic and non-periodic variability by selecting a multiplanet system and flaring stars. We also measure the characteristic variability and flare properties (including microflares) and compare them with prior measurements to assess the impact of our detrending method. Lastly, we discuss the challenges and benefits of this novel detrending technique.