In this work we used a model consisting of a stellar continuum and gas clouds behaving like sticky particles to examine the behavior of gas in transient density waves. In contrast to the prior work of Jog (1992) that employed a gas-dynamical approach we find that the evolution of the density perturbation in the gas remains closely coupled to the one in the stars even when self gravity of the ISM is taken into account.
In a first phase, the ISM responds to the growing potential perturbation by flowing towards the wave crest. When the potential perturbation already decreases, the amplitude in the gas continues to grow until the dissolution of the gaseous arm begins due to Coriolis forces. In this process, a variation of a cloud's epicyclic frequency with its distance from the wave crest leads to a double wave in the velocity perpendicular to the arm, with particles near the wave crest already flowing outward while those further out are still moving in. We regard this as a strong signature for transient density waves that might be within reach of current observational techniques. Finally, the gaseous arms continue to evolve as kinematical spiral arms until they are finally damped out by phase mixing.
For the longer-term evolution we propose that swing amplification events follow one another in a more or less random fashion. We find that for a wide range of excitation frequencies spiral arms continue to form and dissolve. By interference between density waves a ragged morphology results. The dissipativity of the gas - unimportant for the short-term dynamics - keeps the velocity dispersion of the clouds low and thus their responsiveness high.
Given the stochastic nature of the processes described here, one cannot predict the appearance of a galaxy based on a knowledge of its basic state using the methods applied in this work. However, since our results are quite insensitive to the choice of parameters of the underlying model, we regard them as fairly generic for transient density waves. Thus, finding a double wave in radial velocity in the line of sight across a spiral arm for some component of low velocity dispersion (e.g., gas or OB-associations) would make a strong point for the existence of transient spiral arms.
© European Southern Observatory (ESO) 1998
Online publication: February 16, 1998