Addressing the Disk Heating Mechanism of the Late-Type Spiral Galaxy NGC 7664

The scattering due molecular clouds, spiral arms, minor mergers, and bar instabilities has been proposed to explain why disk stars move onto eccentric and inclined orbits, although they formed in a thin layer of gas clouds rotating onto nearly circular orbits in the galactic plane. The shape of the stellar velocity ellipsoid depends on the disk heating mechanism. We constrained the disk heating mechanism of the late-type spiral NGC 7664. The shape of ts velocity ellipsoid was already measured with the asymmetric-drift dynamical modelling of the stellar and ionized-gas kinematics along the major and minor axes of the galaxy. The resulting value of σz /σR = 1.37 ± 0.07 is peculiar and hard to be theoretically explained. We fitted the NGC 7664 i-band image taken from the SDSS archive with a multi-Gaussian expansion model to parametrize the galaxy surface-brightness distribution. We then used this photometric model to fit the stellar kinematics with the Jeans anisotropic model. We found a best-fitting value of σz /σR = 1.30 ± 0.07 by adopting a self-consistent model and σz /σR = 0.97 ± 0.03 by adding a dark matter halo to the model. The result of the self-consistent one is consistent with the one previously found, but the model does not properly fit the kinematic data. On the other hand, the implementation of a dark matter halo significantly improves the best-fitting of the stellar kinematics. We conclude that minor mergers are responsible for the nearly-isotropic stellar velocity ellipsoid of NGC 7664. Indeed, the galaxy shows a very low content of molecular gas and displays a tidally-disturbed morphology of its outermost spiral arm. Our results increase the sample of galaxies with a measured stellar velocity ellipsoid shape. In literature we found a number of galaxies which are consistent with σz /σR > 1, but all the σz /σR measurements were carried out without taking into account the presence of a dark matter halo. This could be the explanation for these peculiar values. However, to achieve ultimate conclusions, a statistical approach, additional data (e.g. integral-field kinematics), and dynamical models that take into account the spatial variations of the stellar velocity ellipsoid shape are needed.