2.5 Dwarf irregular galaxies

Dwarf irregular galaxies can be considered extreme late-type spirals, at least as concerns the rotation curve and the associated dark matter. Many of them present well defined rotation curves that can be obtained from HI measurements. With respect to normal spirals, dwarf irregulars have the advantage of presenting a large gaseous component, and so rotation curves can be traced better and to much larger radii, up to 17 radial scale lengths. Indeed, the rim of the halo has probably been detected (Ashman, 1992).

Kerr, Hindman and Robinson (1954) and Kerr and de Vaucouleurs (1955) showed that the LMC and the SMC were rotating. This result was extended to other irregulars showing that most late-type dwarf galaxies rotate, although their rotation velocities are lower, of the order of $\sim$ 60kms^-1. It was also established that the rotation curve rose slowly to the last measured point. These galaxies were soon considered ideal to study galactic dark matter, not only because the absence of a bulge made analysis simpler, but mainly because the rising curves greatly differed from the expected Keplerian decline. First results (Carignan, 1985; Carignan, Sancisi and van Albada, 1988) seemed to indicate that these galaxies have dark matter properties similar to those of normal spirals and that the inner parts do not require great amounts of dark matter. This trend was not confirmed later, and the commonly accepted picture was that the contribution of the disk is insignificant and that they are dominated by dark matter at all radii. For more details of this history see the thesis by Swaters (1999).

But this conclusion was based on a very small sample of galaxies. The studies by Broeils (1992) and Coté (1995) were based on only eight late-type dwarf galaxies, a small number taking into account the large spread of dark matter properties that this type of galaxies presents. Recently, Swaters (1999) has carried out the greatest effort made to date to systematically observe and analyze this problem. It is also important to have a large sample observed and reduced with the same techniques. To determine the dark matter amounts, it is necessary to obtain photometric maps. In this study, this was done for 171 galaxies at the 2.5m INT at La Palma. Of these, 73 were observed in HI with the Westerbrok Synthesis Radio Telescope. Rotation curves were obtained for 60 of them, and detailed dark matter models were carried out for 35. Clearly, the results obtained in this work are based on the largest and most homogeneous sample. In general, these results did not confirm the previous widely accepted picture; late-type dwarfs are not essentially different from normal brighter spirals, which is more in agreement with the pioneering interpretations.

Despite their apparent loss of symmetry, the exponential decline of typical disks is usually observed in these HI rich galaxies. Swaters found that the rotation curves flatten after about two disk scale lengths. There are several galaxies with fairly flat rotation curves with amplitudes as low as 60 kms^-1. The main difference with rotation curves of spiral galaxies is that no cases of declining curves were found, which was explained by the fact that these galaxies have no bulge at all or only a small one while bright spirals with declining curves do have a large bulge.

The outer slope as a function of R-magnitude is plotted in figure 9 and includes both bright galaxies (from Broeils, 1992) and galaxies belonging to the Ursa Major cluster (Verheijen, 1997). It is seen that the variation in slopes is larger in late-type spirals.

Figure 9. Logarithmic slope between two and three disk scale lengths S_(2.3)^h versus absolute R-band magnitude M_R. Filled circles correspond to late-type galaxies in a high quality rotation curve sample, open circles represent dwarfs in a lower quality rotation curve sample, open triangles are galaxies in Verheijen's (1997) Ursa Major sample, filled triangles represent the galaxies from various sources presented in Broeils (1992). From Swaters (1999) PhD thesis.

In Fig. 10 we reproduce the Tully-Fisher relation from Swaters (1999) that extends the relation to fainter types. We observe L $\propto$ V_max^$\alpha$, where $\alpha$ $\sim$ 4.4. It is also observed that late-type dwarfs rotate noticeably faster than predicted by the Tully-Fisher relation.

Figure 10. The Tully-Fisher relation for spiral and late-type dwarf galaxies. Symbol coding as in Fig. 13. From Swaters (1999) PhD thesis.

In general, late-type galaxies are not dominated by dark matter within the optical disk for radii less than about four scale lengths. The required stellar mass-to-light ratio is however greater than in bright spirals, of the order of 10, reaching values as high as 15. Maximum disk models fit the obtained rotation curves reasonably well, but other models cannot be excluded.

Many irregulars are satellites of bright galaxies or at least of a small group of galaxies as in the Local Group. Here, a gradation in the properties from dEs to dIrrs would support the hypothesis that Irrs could eventually evolve into dEs (e.g. Aparicio et al. 1997; Martinez-Delgado 1999). Phoenix could be a clear example of an intermediate type. Moreover, dEs are preferentially distributed close to the largest galaxies, while dIrrs are found in the outskirts (the Magellanic Clouds are exceptions) (van den Bergh 1999).

Salucci and Persic (1997) proposed that the "universal rotation curve" was also valid for dwarf irregulars, though then the large amount of data in the thesis of Swaters was not available. In this case, the calculation of a and $\beta$ is given by different formulae:

	(18)
	(19)

if

(20)

Thus, for a bright dwarf irregular, L $\sim$ 0.04L_*, V_opt $\sim$ 63kms^-1, characteristic values are a = 0.93 and $\beta$=0.08. Therefore the core radius is nearly as large as the optical radius and the contribution of the visible matter at the optical radius is nearly negligible. If L < 0.04L_*, $\beta$ is even lower and a higher. Under this interpretation, therefore, dwarf irregulars are very dark galaxies, have very dense halos and large masses, obtainable with 8 × 10¹⁰(L/0.04L_*)^1/3.

Salucci and Persic (1997) give a formula to estimate the total mass of a galaxy as a function of its visible mass

(21)

Thus, when M_visible is large the M₂₀₀/M_visible ratio decreases. Merely to state that M_total is proportional to L, as is often done, would be to make a bad assumption, worse than supposing that all galaxies have equal mass, irrespective of their luminosity.

Though most studies of these galaxies conclude that moderate or large amounts of dark matter are required, we cannot exclude the magnetic interpretation of the data for spiral galaxies, which should also be taken into account for dwarf irregulars. Under this interpretation, the higher magnetic fields required imply larger escaping fluxes that are actually observed, for instance in M82 (also associated with the ejection of magnetic fields (Reuter et al., 1992; Kronberg and Lesch, 1997) or in NGC 1705 (Meurer, Staveley-Smith and Killeen, 1998), a galaxy requiring specially high DM central density (0.1 M_$\odot$pc^-3) and a large mass loss rate of the order of 0.2-2 M_$\odot$yr^-1.