Exploring individual binaries and a stochastic background in population-based simulated pulsar timing array datasets

The dominant source of gravitational waves in the nHz frequency band is expected to be a population of supermassive black hole binaries. The vast majority of these binaries are too weak to be individually detected, however they result in the emergence of a stochastic gravitational wave background. Ongoing pulsar timing array experiments are expected to reach sufficient sensitivities to detect such a stochastic signal within the next few years. As more data is collected, we will also be able to find individual supermassive black hole binaries, which will be excellent sources for multimessenger studies. To understand the challanges of detecting and characterizing either of these signals in a realistic setting, we create simulated datasets where we individually add the signals of hundreds of millions of binaries. For this we use synthetic catalogs of binaries based on the Illustris cosmological hydrodynamic simulation. We characterize deviations from the simple isotropic Gaussian background model due to the finite number of contributing sources, and we explore how that affects the detectability and parameter estimation of individual binaries and the stochastic background.