Age velocity dispersion relations and heating histories in disc galaxies

Abstract: We analyse the heating of stellar discs by non axisymmetric structures and giant molecular clouds (GMCs) in N-body simulations of growing disc galaxies. The analysis resolves long-standing discrepancies between models and data by demonstrating the importance of distinguishing between measured age-velocity dispersion relations (AVRs) and the heating histories of the stars that make up the AVR. We fit both AVRs and heating histories with formulae proportional to t^beta and determine the exponents beta_R and beta_z derived from in-plane and vertical AVRs and ~beta_R and ~beta_z from heating histories. Values of beta_z are in almost all simulations larger than values of ~beta_z, whereas values of beta_R are similar to or mildly larger than values of ~beta_R. Moreover, values of beta_z (~beta_z) are generally larger than values of beta_R (~beta_R). The dominant cause of these relations is the decline over the life of the disc in importance of GMCs as heating agents relative to spiral structure and the bar. We examine how age errors and biases in solar neighbourhood surveys influence the measured AVR: they tend to decrease beta values by smearing out ages and thus measured dispersions. We compare AVRs and velocity ellipsoid shapes sigma_z/sigma_R from simulations to Solar neighbourhood data. We conclude that for the expected disc mass and dark halo structure, combined GMC and spiral/bar heating can explain the AVR of the Galactic thin disc. Strong departures of the disc mass or the dark halo structure from expectation spoil fits to the data.