Modeling Variability in Black Hole Images
The Event Horizon Telescope (EHT), a mm-wavelength very long baseline interferometer (VLBI), aims to take the first ever resolved images of black holes at event horizon scales. Interferometers detect the complex Fourier components (visibilities) of a source and use image synthesis algorithms to reconstruct the image. A fundamental assumption of most image reconstruction algorithms is that the source remains stationary throughout the length of the observation. A primary target for the EHT is the Galactic center black hole Sagittarius A* (Sgr A*), which is known to be variable on timescales significantly shorter than the observational timescale for the EHT. We use magnetohydrodynamic simulations and radiative transfer calculations in general relativity to characterize the effect of intrinsic source variability on interferometric observables. We show that intrinsic source variability will significantly affect conventional image reconstruction techniques and that variability must be taken into account for both image synthesis and model fitting. Furthermore, we explore the utility of Principal Component Analysis (PCA) to characterize the structural variability in GRMHD simulations of Sgr A* and find that simulations can be compactly represented with a PCA-derived basis of eigenimages. This allows for detailed comparisons between variable observations and time-dependent models.