Numerical simulations of bar formation in the Local Group

TL;DR

This study uses cosmological hydrodynamical simulations to investigate the formation and evolution of stellar bars in galaxies within the Local Group, finding that simulated bars are shorter and faster than previous models, aligning better with observations.

Contribution

It presents new cosmological simulations that produce shorter, faster-rotating bars, reducing the discrepancy between theory and observations in bar properties.

Findings

Simulated bars are shorter and rotate faster than previous cosmological models.

Bars form in systems where disc self-gravity dominates over dark matter.

Bars grow over 3-4 Gyr, increasing in length while slowing rotation, matching observed properties.

Abstract

More than 50 per cent of present-day massive disc galaxies show a rotating stellar bar. Their formation and dynamics have been widely studied both numerically and observationally. Although numerical simulations in the $\Lambda$CDM cosmological framework predict the formation of such stellar components, there seems to be a tension between theoretical and observational results. Simulated bars are typically larger in size and have slower pattern speed than observed ones. We study the formation and evolution of barred galaxies, using two $\Lambda$CDM zoom-in hydrodynamical simulations of the CLUES project that follow the evolution of a cosmological Local Group-like volume. We found that our simulated bars, at $z = 0$, are both shorter and faster rotators than previous ones found in other studies on cosmological simulations alleviating the tension mentioned above. These bars match the short tail-end of the observed bar length distribution. In agreement with previous numerical works, we find that bars form in those systems where the disc self-gravity is dominant over the dark matter halo, making them unstable against bar formation. Our bars developed in the last 3-4 Gyr until they achieve their current length and strength; as bars grow, their lengths increase while their rotation speeds decrease. Despite this slowdown, at redshift $z = 0$ their rotation speeds and size match well the observational data.