Astron. Astrophys. 349, 411-423 (1999)

3. Results

The complete HI data cube is shown in Fig. 2. The flux added over all channels with a cutoff of 2 is shown in Fig. 3. The total flux calculated using a cutoff of 2 is = 9.4 Jy km s^-1. This corresponds to a total HI mass of with a distance D = 17 Mpc for the Virgo cluster ³. This is in good agreement with the value given by Cayatte et al. (1990) using a larger beam size of .

Fig. 2. The HI channel maps. The heliocentric velocity relative to the reference velocity is indicated in km s-1 in the upper right of each channel. The contour lines are 1.3, 2.6, 3.9, 5.2, and 6.5 mJy/beam.

Fig. 3. The optical B image together with the HI contour map of NGC 4548. The contour levels correspond to 0.82, 3.24, 6.48, 9.73, 12.97, 16.21, 19.45 cm-2. The beam is shown in the lower left corner. In order to distinguish maxima and minima we refer to Fig. 5.

The HI emission is distributed within an almost complete ring. The maximum of emission is located in the south-east. The emission profile along the minor axis is quite symmetrical, whereas the one along the major axis is strongly asymmetrical. In fact the south-eastern emission maximum has no counterpart in the north-west. The near Infrared image of Boselli et al. (1997b Fig. 3b) shows clearly the bar and the two spiral arms in the north-west and south-east. As for the HI emission, there is more NIR emission coming from the southern part of the galaxy. In addition, the outer limits of the NIR emission coincide well with the one of the HI emission. Thus, the intensity of the HI emission follows the gravitational potential traced by the NIR image. In the galaxy's centre the HI emission drops by an order of magnitude leading to an in east-west direction elongated hole. It is also worth noticing that the inner edge of the emission ring extends more inwards in the north than in the south.

In order to compare the galaxy's gas content to its stellar population, we show the HI emission together with an optical B image in Fig. 3. One clearly recognizes the bar which ends at the inner edge of the HI ring. It is also visible that both the low contrast spiral arms are traced by the HI emission. The local HI emission maximum in the extended southern arm is associated with the young stellar population traced by the spiral arm. The outer edge of the atomic gas ring follows exactly the shape of the stellar disc. A dust lane is seen in absorption near the centre in the south-west. If one accepts the idea that it is not located in the inner disc, this indicates that the eastern side is the near side of the galaxy.

As expected for an anemic spiral galaxy the H line map shows very few HII regions. In Fig. 4 we show this map together with the HI emission map. The most luminous HII regions are located along the beginning of the spiral arms at the end of the bar. There the interaction between the bar and the outer gas favours star formation.

Fig. 4. Contour plot of the HI emission map together with the H image. The contour levels correspond to column densities of 3.44, 6.89, 10.33, 13.78, 17.22, 19.291020 cm-2. The HI beam is shown in the lower left corner.

The CO emission was observed in a region centered on the galaxy. These data together with the HI map are shown in Fig. 5. As expected, the bar appears clearly in the CO emission. We can also observe the points where the local CO emission maxima join the HI emission maxima at the ends of the bar.

Fig. 5. The HI emission map from Fig. 3 together with CO emission contour map. The maximum level is 10.87 K km s-1. The contour levels are in steps of 1.09 K km s-1. The HI beam is shown in the lower left corner.

Honma et al. 1995 demonstrated that the gas phase transition between HI and H₂ occurs within a small radial distance. The fraction of H₂ to the total gas column density (molecular fraction) increases very rapidly inwards within this boundary region. In order to study this effect for NGC 4548, the deprojected distance for each CO pointing to the galaxy centre was calculated with the help of the position angle and the inclination (see next section). As the HI and CO data have similar beam sizes it is possible to compare the column densities at a given position. The fraction of column densities for each CO pointing is plotted as a function of the deprojected distance (Fig. 6). We assumed CO conversion factor of cm^-2 (K km s as derived by the EGRET gamma-ray observations (Digel et al. 1996) for the solar neighbourhood. However, in the Perseus arm at 3-4 kpc from the Sun cm^-2 (K km s (Digel et al. 1996). For extragalactic sources there are only estimations ranging from cm^-2 (K km s for M51 (Guélin et al. 1995) to cm^-2 (K km s for the SMC (Lequeux et al. 1994). In general it seems that X does not differ by a large factor in spiral galaxies with a luminosity similar to the Galaxy (Boselli et al. 1997a). Therefore, we have adopted a factor 3 for the uncertainties in the determination of X. The error bars in Fig. 6 represent these uncertainties. As there is no HI detection in the centre there. The molecular fraction tends to decrease with radius up to which corresponds approximately to the radial extent of the bar. There the predominantly atomic gas appears to be transformed into molecules due to the compression caused by the bar. Further out the molecular fraction does not show a further decline. This means that we observe a sharp transition between molecular and atomic gas at about 30" and a constant molecular gas fraction (assuming cm^-2 (K km s^-1)^-1) further out.

Fig. 6. The molecular fraction at each CO pointing. Its values as a function of the deprojected distance to the galaxy centre are shown. The error bars correspond to conversion factors of and X3.

© European Southern Observatory (ESO) 1999

Online publication: September 2, 1999
helpdesk.link@springer.de