State of the Universe

In pursuit of primordial non-Gaussianity

by Prof. Daniel Meerburg (University of Groningen)

Friday, September 18, 2020 from to (Asia/Kolkata)
at Zoom Meeting ID: 838 9995 1661 Passcode: 420478
In this talk I will motivate the search for primordial non-Gaussianity. Current constraints derived from studying the bispectrum of the cosmic microwave background (CMB) suggest a Gaussian primordial Universe. However, experimental efforts this decade are expected to reach a sensitivity that would reach a compelling, theoretically motivated, threshold. I will discuss both fundamental as well as observational limitations that affect our ability to constrain or detect a primordial signal. In general, it has always been assumed the most sensitive statistic of primordial non-Gaussianity is the bispectrum (the Fourier transform of the 3-point function). I will show that the signal-to-noise (SNR) scaling of ANY n-point correlation can be determined by i) the kinematics of the n-point function is certain critical configurations and ii) the presence or absence of damping (e.g. Silk damping in the CMB). As a result, I can show that the SNR of the CMB bispectrum scales poorly with increased resolution, while the trispectrum (Fourier equivalent of the 4-point function) in certain cases can improve over bispectrum sensitivity as experimental resolution increases.. While CMB sensitivity to higher n-point functions will eventually plateau as a function of resolution, for the large scale structure, which is 3 dimensional and structure is not fundamentally damped on small scales, all n-point functions can in principle improve over bispectrum constraints with increasing resolution. This unknown results motivates a continued effort to search for primordial non-Gaussianty. If time permits, I will also discuss recent work on using a single tracer field of the LSS to measure non-Gaussianity in the biased power spectrum, utilizing so-called cosmic variance cancellation.