Astron. Astrophys. 335, 421-430 (1998)

7. Gradients

For galaxies F563-1, UGC 5005 and UGC 5999 we have enough measurements over a large enough radial range that we may attempt to investigate possible radial oxygen abundance gradients. We have plotted in Fig. 6 the abundances of the HII regions in these galaxies versus the deprojected radial distance, expressed in scale lengths. There is no clear trend of abundance with radius. Rather the oxygen abundance seems to be constant at .

Fig. 6. Oxygen abundances in HII regions of 3 LSB galaxies. The true radial distance to the center (corrected for inclination effects) is expressed in optical scale lengths. The circles denote galaxy F563-1; the squares denote UGC 5005 and the triangles UGC 5999. No strong trend is seen. HSB late-type galaxies show a gradient of dex in (O/H) over the radial range shown here.

To compare with other galaxies of similar Hubble type, we show in Fig. 6 the abundances of HII regions in 9 late-type Scd-Irr galaxies, taken from VCE (their Fig. 2). We have converted their half light radii to disk scale lengths using the relation , under the implicit assumption that the light distribution of their sample galaxies can be described by an exponential disk. As we are dealing with very late-type galaxies this assumption is most likely justified.

The VCE galaxies show a gradient of dex over the range in radius sampled by the LSB galaxies. The LSB HII regions clearly do not follow this trend. The oxygen abundances in the LSB galaxies are lower at all radii, except in the very outer parts where the abundance values of the HSB and LSB galaxies appear to converge. The lack of abundance gradients in LSB galaxies may indicate that the picture of galaxies evolving from the inside out may not apply to LSB galaxies. It is usually assumed that the abundance and colour gradients found in galaxies indicate that the outer parts of galaxies are less evolved than the inner parts. The colours and abundances in LSB galaxies are comparable with those in the outer (most unevolved) parts of HSB galaxies. The absence of a radial abundance trend in the LSB galaxies suggests that these may have evolved at the same rate over their entire disk. Indeed, a spatially and temporally sporadic star formation rate, as is derived for LSB galaxies (e.g. Gerritsen & de Blok 1997), would not give rise to an abundance gradient.

Alternative explanations for the lack of a gradient may be the infall of metal-poor gas from high above the planes of these galaxies. However, because of the low star formation activity in LSB galaxies it is not likely that large amounts of gas can have been blown out in the past. If gas infall is a major factor, this must be "primordial" gas, left over from when the LSB galaxies were formed.

A second alternative is that the disks of LSB galaxies are still settling in their final configuration. Gas from larger radii is slowly diffusing inward, causing density enhancements where conditions for star formation may be favourable. This is qualitatively consistent with the finding that most of the star forming regions in LSB galaxies are found towards the outer radii of the stellar disk. If star formation in LSB galaxies needs to be stimulated by external conditions (like infall), and is not self-propagating, this building up of the disk would not give rise to an abundance gradient. If every radius would go through one cycle of star formation before fading, after which star formation would move on to larger radii, this would cause a colour gradient (as is observed) but no abundance gradient. Note that though this evolution "from the inside out" is different from the standard picture mentioned above. In that picture each radius continues evolving, with the inner radii going through more star formation cycles than the other radii. Clearly, the scenario described here is only a crude ad-hoc attempt to explain the lack of gradient. It would have to be tested by modelling, and supported by more abundance data.

In this respect the discussion presented in Edmunds & Roy (1993) is of interest. They show that steep abundance gradients in gas-rich disk galaxies seem to require the presence of unbarred spiral structure. The abundance gradients disappear at the same absolute magnitude that spiral structure ceases () and are considerably shallower in galaxies with a strong bar.

The late-type LSB galaxies in general show only a faint spiral structure, and have . Their feeble spiral structure may thus be related with the lack of abundance gradients. Edmunds & Roy (1993) offer two possible explanations. The variation with radius of the frequencey with which interstellar material passes through a spiral pattern may result in a declining star formation rate with radius. This would set up an abundance gradient. At lower absolute magnitudes star formation will still continue, but the spiral pattern would be absent. This would result in a star formation rate which is much more uniform across the disk than in strong spirals. Alternatively, if non-circular motions are relatively more important the resulting mixing may also homogenize the chemical composition of the interstellar medium.

Although we realize that it is easy to over-interpret data of such a limited dynamic range, we conclude that the lack of abundance gradients confirms the evolutionary picture of quietly evolving LSB galaxies with only local processes regulating their evolution.

© European Southern Observatory (ESO) 1998

Online publication: June 18, 1998
helpdesk.link@springer.de