3.1. Inner slope values : technical and selection issues

Most observers conclude that in LSB galaxies is closer to 0.0 than to -1.0, and almost all find that the decomposition mass models work better if the dark halo is modelled with a (pseudo-) isothermal sphere, or a Burkert profile, rather than with the NFW profile. However, some authors maintain that the problem of determining the inner slope is fraught with systematic effects, which all magically contribute towards shallower slopes. The issue debated is not only about the best value for the inner slope, but also whether a slope of -1.0 can be reasonably excluded, so as to force a modification of the CDM picture at galaxy scales. See also the contributions by De Blok and by Swaters (this volume).

On the selection side, the debate is about the quality and relevance of the data. As more galaxies are observed, stricter selection criteria can be used to retain only galaxies with sufficiently "good" data. Thus one excludes 1) galaxies with poor angular resolution, thereby requiring in most cases supplementary H data in addition to a HI rotation curve, so that the inner 1 kpc of a galaxy is well probed. This excludes far away galaxies or galaxies for which there is only low resolution HI data, and thus minimizes slit width effects; 2) edge-on galaxies, which are apparently too difficult to understand for some workers in the field, despite demonstrations by Bosma et al. (1992) and Matthews & Wood (2001) that small galaxies seen edge-on are transparent and have rotation curves which are slowly rising in the inner parts. The argument that somehow there is no emission at the tangent point leads to very peculiar H distributions in more face-on galaxies which are yet to be seen; 3) galaxies with large asymmetries, faint emission, etc., which have low quality rotation curves.

De Blok, Bosma & McGaugh (2003) applied these criteria to data of McGaugh et al. (2001) and De Blok & Bosma (2002), and find for a restricted sample that the values of the inner slope are between 0.2 and -1.0 (see De Blok, this volume). Swaters et al. (2003a), in a similar study, come to a similar conclusion. De Blok et al. (2003) also report about deliberate slit offsets and misalignments for one galaxy, UGC 4325, and conclude that these effects could not magically bring the slope from about -0.2 to -1.0, although of course scatter is introduced if the major axis is not observed correctly.

Finally, for several galaxies there are more than one observation, done with different telescopes, and by different observers. Even though the error bars are sometimes unsatisfactory large, there is good agreement in general. The situation for UGC 4325, debated at the meeting, is shown in Figure 3. The HI position-velocity diagram prompts me to think that the peak velocity is indeed around ~ 110 km/s as it is in the optical, after which the rotation curve declines.

Figure 3. Comparison of data for UGC 4325. HI data from Swaters 1999 (HI), Fabry-Pérot data from Garrido et al. 2002 (FP), long slit data from Marchesini et al. 2002 (Mar), De Blok & Bosma 2002 (B02), De Blok et al. 2003 (B03), and Swaters et al. 2003a (S03). Bsm represents smoothed long slit data as described in B02. At bottom the HI distribution and position-velocity diagram from Swaters 1999.