We present hydrodynamical models of the grand design spiral M51 (NGC 5194), and its interaction with its companion NGC 5195. Despite the simplicity of our models, our simulations capture the present‐day spiral structure of M51 remarkably well, and even reproduce details such as a kink along one spiral arm, and spiral arm bifurcations. We investigate the offset between the stellar and gaseous spiral arms, and find at most times (including the present day) there is no offset between the stars and gas within our error bars. We also compare our simulations with recent observational analysis of M51. We compute the pattern speed versus radius, and similar to observations, find no single global pattern speed. We also show that the spiral arms cannot be fitted well by logarithmic spirals. We interpret these findings as evidence that M51 does not exhibit a quasi‐steady density wave, as would be predicted by density wave theory. The internal structure of M51 derives from the complicated and dynamical interaction with its companion, resulting in spiral arms showing considerable structure in the form of short‐lived kinks and bifurcations. Rather than trying to model such galaxies in terms of global spiral modes with fixed pattern speeds, it is more realistic to start from a picture in which the spiral arms, while not being simple material arms, are the result of tidally induced kinematic density ‘waves’ or density patterns, which wind up slowly over time.

Here we present our two-dimensional chemodynamical code CoDEx, which we developed for the purpose of modeling the evolution of galaxies in a self-consistent manner. The code solves the hydrodynamical and momentum equations for three stellar components and the multiphase interstellar medium (clouds and intercloud medium), including star formation, Type I and Type II supernovae, planetary nebulae, stellar winds, evaporation and condensation, drag, cloud collisions, heating and cooling, and stellar nucleosynthesis. These processes are treated simultaneously, coupling a large range in temporal and spatial scales, to account for feedback and self-regulation processes, which play an extraordinarily important role in the galactic evolution. The evolution of galaxies of di †erent masses and angular momenta is followed through all stages from the initial protogalactic clouds until now. In this Ðrst paper we present a representative model of the Milky Way and compare it with observations. The capability of chemodynamical models is convincingly proved by the excellent agreement with various observations. In addition, well-known problems (the G-dwarf problem, the discrepancy between local e †ective yields, etc.), which so far could be only explained by artiÐcial constraints, are also solved in the global scenario. Starting from a rotating protogalactic gas cloud in virial equilibrium, which collapses owing to dissipative cloud-cloud collisions, we can follow the galactic evolution in detail. Owing to the collapse, the gas density increases, stars are forming, and the Ðrst Type II supernovae explode. The collapse time is 1 order of magnitude longer than the dynamical free-fall time because of the energy release by Type II supernovae. The supernovae also drive hot metal-rich gas ejected from massive stars into the halo, and as a consequence, the clouds in the star-forming regions have lower metallicities than the clouds in the halo. The observed negative metallicity gradients do not form before t \ 6 ] 109 yr. These outward gas Ñows prevent any clear correlation between local star formation rate and enrichment and also prevent a unique age-metallicity relation. The situation, however, is even more complicated, because the mass return of intermediate-mass stars (Type I supernovae and planetary nebulae) is delayed depending on the type of precursor. Since our chemodynamical model includes all these processes, we can calculate, e.g., the [O/H] distribution of stars and Ðnd good agreement everywhere in bulge, disk, and halo. From the galactic oxygen to iron ratio, we can determine the supernovae ([II ] Ib]/Ia) ratio for di †erent types of Type Ia supernovae (such as carbon deÑagration or sub-Chandrasekhar models) and Ðnd that the ratio should be in the range 1.0È3.8. The chemodynamical model also traces other chemical elements (e.g., N ] C), density distributions, gas Ñows, velocity dispersions of the stars and clouds, star formation, planetary nebula rates, cloud collision, condensation and evaporation rates, and the coo...

Gravitational tides are widely understood to strip and destroy galactic substructures. In the course of a galaxy merger, however, transient totally compressive tides may develop and prevent star forming regions from dissolving, after they condensed to form clusters of stars. We study the statistics of such compressive modes in an N-body model of the galaxy merger NGC 4038/39 (the Antennae) and show that ~15% of the disc material undergoes compressive tides at pericentre. The spatial distribution of observed young clusters in the overlap and nuclear regions of the Antennae matches surprisingly well the location of compressive tides obtained from simulation data. Furthermore, the statistics of time intervals spent by individual particles embedded in a compressive tide yields a log-normal distribution of characteristic time ~10 Myr, comparable to star cluster formation timescales. We argue that this generic process is operative in galaxy mergers at all redshifts and possibly enhances the formation of star clusters. We show with a model calculation that this process will prevent the dissolution of a star cluster during the formation phase, even for a star formation efficiency as low as ~10%. The transient nature of compressive tides implies that clusters may dissolve rapidly once the tidal field switches to the usual disruptive mode.Comment: 5 pages, 3 figures, accepted for publication in MNRAS Letters. For higher resolution, see http://astro.u-strasbg.fr/~renaud/publi/mnras08.pd

Abstract. NGC 4449 is an active star-forming dwarf galaxy of Magellanic type. From radio observations, van Woerden et al. (1975) found an extended HI-halo around NGC 4449 which is at least a factor of 10 larger than the optical diameter D25 ≈ 5.6 kpc. Recently, Hunter et al. (1998) discerned details in the HI-halo: a disc-like feature around the center of NGC 4449 and a lopsided arm structure. We combined several N-body methods in order to investigate the interaction scenario between NGC 4449 and DDO 125, a close companion in projected space. In a first step fast restricted N-body models are used to confine a region in parameter space reproducing the main observational features. In a second step a genetic algorithm is applied for a uniqueness test of our preferred parameter set. We show that our genetic algorithm reliably recovers orbital parameters, provided that the data are sufficiently accurate, i.e. all the key features are included. In the third step the results of the restricted N-body models are compared with self-consistent N-body simulations. In the case of NGC 4449, the applicability of the simple restricted N-body calculations is demonstrated. Additionally, it is shown that the HI gas can be modeled here by a purely stellar dynamical approach. In a series of simulations, we demonstrate that the observed features of the extended HI disc can be explained by a gravitational interaction between NGC 4449 and DDO 125. According to these calculations the closest approach between both galaxies happened ∼4 − 6 10 8 yr ago at a minimum distance of ∼25 kpc on a parabolic or slightly elliptic orbit. In the case of an encounter scenario, the dynamical mass of DDO 125 should not be smaller than 10% of NGC 4449's mass. Before the encounter, the observed HI gas was arranged in a disc with a radius of 35-40 kpc around the center of NGC 4449. It had the same orientation as the central ellipsoidal HI structure. The origin of this disc is still unclear, but it might have been caused by a previous interaction.

Abstract.Here we present as an extension to already published models of massive spherical galaxies one-dimensional chemodynamical models of dwarf elliptical galaxies in the mass range between 10 9 and 10 10 M and initial density fluctuation of 1σ and 3σ. Because of their vanishing angular momentum the models are restricted to spherical symmetry. Due to this limitation the dynamics of the different components consider only radial motions and also their interaction processes have only one degree of freedom. Therefore, the galaxy evolution is preferably determined by radial oscillatory phenomena caused by heating and cooling of the interstellar gas, which reenforce effects like starbursts. Nevertheless, the low gravitational binding energy of dwarf ellipticals can easily be exceeded by thermal and turbulent energy production in the interstellar medium, leading to gas expansion and even to galactic winds. Furthermore, gas phases as well as gas and stars can decouple dynamically during the galactic evolution. For comparison with observational signatures like stellar kinematics, radial densities and metallicity distributions of the different components, the chemo-dynamical treatment of galaxy evolution must take the multi-phase character of the interstellar medium and self-regulation of star formation at low gravitation into account. Since non-rotating low-mass galaxies with stochastic star-formation episodes are seen as dwarf ellipticals, the aim of this paper is to compare the 1d chemo-dynamical models with observed characteristics of this type of dwarf galaxies. Several features like stellar populations from separate formation episodes, low metallicities, significant mass loss, etc. can be reproduced by the models in a global manner. Quantitative disagreements between model predictions and observations provide insight into necessary improvements of subsequent chemo-dynamical models implying rotation, dark matter halos, and external effects by an intergalactic gas, like its pressure, gas infall and stripping of galactic gas. The results can be summarized as follows: 10 9 M models are characterised by a single star-formation event during the initial collapse with an active self-regulation leading to reexpansion and massive gas loss as well as remaining at very low metallicities; 10 10 M dwarf elliptical models become almost gas-free due to large initial star formation and a strong galactic wind. These as well as a 5 × 10 9 M and 3σ model experience oscillatory but non-negligible later star-formation histories that can account for the observed intermediate-age stellar populations.