We presented a theoretical model for the regular magnetic field in NGC 6946 based on the concept of a turbulent dynamo and we confronted our numerical results with present observations of polarized radio emission.
Our dynamo model assumes axisymmetric distributions of large-scale velocity, density contribution and turbulence intensity, because observations indicate only weak azimuthal variations. The optical spiral arms were introduced only via the correlation time of interstellar turbulence. The radial distribution of the gas density was also taken from observations. Since there is some discord in the literature about the rotation in NGC 6946 we investigated two possible rotation curves (Carignan et al. 1990; Sofue 1996). We further introduced a vertical stratification of the gas density and of the turbulence intensity. Without accurate knowledge about the distribution in NGC 6946 we assumed a stratification similar to that in our own Galaxy. Our dynamo model is nonlinear: We allowed the back-reaction of magnetic field onto the -effect, but we neglected the influence onto the eddy diffusivity and on the large-scale gas motion.
Our basic results can be summarized as follows:
(i) Rotation curve. The observed narrow distribution of polarized radio intensity (cf. Fig. 17) is well reproduced by assuming equipartition between the cosmic-ray and magnetic field energy densities and a rotation curve similar to that one adopted by Sofue (1996). A model using the rotation curve of Carignan et al. (1990) leads to a too broad distribution of the regular magnetic field. However, our model fails within 1 kpc radius because the inner part of the observed rotation curve could not be modeled appropriately.
(ii) Radial dependency of turbulence intensity. There are some observational indications of an outward-decreasing turbulence intensity (Boulanger & Viallefond 1992). Although this decrease can reproduce the observed outwards decreasing absolute value of the magnetic pitch angle, it leads to considerable changes in the properties of the model (field reversals, strong vertical field components) which are in conflict with Faraday rotation data. For our `best model' based on Sofue's rotation curve we chose a radially constant and rather large turbulence intensity (15 km s-1 mid-plane value) to achieve magnetic pitch angles of the right order (cf. Fig. 18). However, we were not able to reproduce its radial decrease satisfactorily.
(iii) Spiral arms. The basic property of the "magnetic arms" between the optical arms are well reproduced in our dynamo models (a result of superposition of magnetic field modes S0 and S2). The arm/interarm contrast in the polarized intensity and the corresponding values of the magnetic pitch angle in agreement with the observations are achieved by a contrast in only the correlation time of interstellar turbulence, varying between 0.02 Gyr (arm) and 0.01 Gyr (interarm) (cf. Fig. 19). The adopted pitch angle of the optical arms has almost no influence onto the magnetic pitch angle in all our models.
(iv) Corotation radius. Our dynamo model for NGC 6946 predicts a phase shift between magnetic and preceding optical arm from 65o to 90o which is similar to the observed values if we assume a corotation radius of about 10 kpc. This contradicts the investigation by Elmegreen et al. (1992). The effect of the corotation radius onto the magnetic field may animate further observations in order to get improved data of the velocity field in NGC 6946.
Future investigations should take into account possibly important effects, e.g. the steep rotation curve at inner radii, the true vertical density stratification, and the magnetic feed-back onto the eddy diffusivity (-quenching) and onto the large-scale gas motion (density waves). New sensitive CO and HI observations should provide improved data on the velocity field, the velocity dispersion and the vertical density distribution of the gas in NGC 6946. Knowledge of the gas density in the magnetic arms requires deep searches for cold gas between the optical arms through CO emission and absorption line observations.
© European Southern Observatory (ESO) 1999
Online publication: October 4, 1999