Detector Noise Correlation Analysis for the Super Cryogenic Dark Matter Search at SNOLAB

Author: Kevin Wykoff, Department of Physics

Mentor: Dr. Richard Schnee, Department of Physics

Well-established astronomical observations indicate that dark matter accounts for approximately 85% of total matter in our universe. Accordingly, obtaining a conclusive dark matter detection is a primary focus in experimental particle astrophysics. One such effort is the current iteration of SuperCDMS (the Super Cryogenic Dark Matter Search) at SNOLAB, an Ontario-based underground laboratory. The experiment will employ cryogenically cooled silicon and germanium crystals, in which miniscule vibrations and charge depositions would indicate a dark matter interaction. The small signal from the interactions expected necessitates specialized electronics and significant detector sensitivity. Critical to ensuring this sensitivity is the extensive characterization and minimization of noise in the detector signals. To study detector noise we acquire time-series of data from the detectors at random times, then use them in the calculation of power spectral densities (PSDs). These PSDs show the variance of noise as a function of frequency. Examination of the PSDs across multiple detectors enables a statistical approach to identifying noise sources. Cross-correlation coefficients are obtained to determine the degree to which noise across multiple detectors is anticorrelated, uncorrelated, and correlated. Utilizing this approach under varying conditions and detection modes, we are able to identify and account for or eliminate noise, potentially increasing the statistical strength of a dark matter detection.