Astron. Astrophys. 344, 421-432 (1999)

4. The photometric and kinematical properties

In Paper I we give the whole set of parameters measured for the sample of isolated galaxies, together with the descriptions of how they have been obtained. In Table 1 we show the median values together with the dispersion and the range of variation for the different measured parameters. Indeed, given the size of the sample it is not possible to give the values for each morphological subtype. Moreover, as we have discussed, most of the galaxies are late types, with 11 of them classified as Sc. We note that this aspect should act as a caution when trying to do comparative studies.

Table 1. Average properties of the 22 isolated spirals in the sample.
Notes:
M are absolute magnitudes. All the R (effective radius) together with a and b (major and minor axes) are in Kpc. D are distances in Mpc (H₀=75 km s^-1 Mpc^-1). i = acos (b/a), in degrees. are surface brightness parameters in mag/(")², and the effective radii in Kpc, from the photometric decomposition, where i is the filter and j is the component (D=disk, B=bulge). O and P are the origin and slope of the linear regime of the color gradients (see text). G is in units of km s^-1 Kpc^-1. M/L_B is in solar units. Velocities are in km s^-1. Mass in 10¹⁰M  units
(1) See text for the definition of = (^o)
(2) Luminosity in 10⁴¹ erg s^-1 units.

4.1. The photometric properties

4.1.1. Luminosity and color indices

The distances used to evaluate the luminosity were derived from the measured redshift corrected for galactocentric motion (as indicated in the RC2 catalogue, de Vaucouleurs et al., 1976), with H₀ = 75 km s^-1 Mpc^-1. The correction for the Virgocentric inflow for our galaxies is in general small (always less than 15%) and, as discussed in Paper I, not sensitive to the detailed model used. We therefore decided not to correct for the inflow. The magnitudes and color indices given in Table 2 are corrected for galactic and internal extinction as explained in Paper I (see Table 7 in Paper I).

The range of the B luminosity of the galaxies we have measured is , with a median value of -20.35 (see Table 1). Therefore, there are no faint spirals in our sample, a fact that has to be taken into account when making comparative analysis (see below). The effective radii in Table 1 have been calculated from the growing curves in the different bands. The mean surface brightness values were evaluated for the area enclosed by the 25 mag/(")² isophote (taken from the RC3), i. e., , where is the radius of that isophote.

For the total color indices the ranges we find are and . These are similar to what is found for other samples of spiral galaxies, independently of their interaction status (Roberts & Haynes 1994; de Jong 1996c).

To characterize the populations of the disks we have considered the color indices and their gradients along the disk. To quantify the color gradients we have fitted a linear function of the form CI(r) = CI(c) + Pr, where CI(c) is the (extrapolated) central color index under consideration, and r the radial distance in kpc. We find that P is negative for all the galaxies in the sample, i.e., their color indices become bluer towards the external parts. As shown in Fig. 1, the central color indices correlate with the corresponding total colors: they are redder for redder galaxies. The correlation is better traced by , with correlation coefficient r = 0.774 and probability P = 0.9969. Another interesting aspect is that the central colors seem to span a wider range that outer colors, specially in (B-I). We find and for the central colors, whereas for the colors measured at the last recorded isophote, R_B, we have and (UGC 3511 was excluded for this calculation since the colors we measured are abnormally red for a Scd spiral, see Paper I).

Fig. 1. The central colors, and the slope P of the color gradient (in magnitudes per kpc) as a function of the total galactic colors.

The emerging picture from the above considerations is that redder galaxies have redder central colors. The fact that the range of color indices at the outer regions is smaller than at the center would mean that the disks of different galaxies tend to be more alike that their bulges.

4.1.2. The properties of the bulges and disks

It is a well known fact that the output of the photometric decomposition of the light distribution in the image of a galaxy depends on the method used and on the form of the profiles adopted for the components (Knapen & van der Kruit 1991). To make explicit our choices we have used 1-D light profiles, with an exponential law for the disk (Freeman 1970) and the law for the bulge (de Vaucouleurs 1948), respectively:

and

where i stands for the photometric band under consideration.

The isophotal profiles have been derived by plotting the isophotal levels versus their equivalent radii, calculated from the area inside each observed isophote. Disk and bulge parameters have been obtained from the surface brightness profiles, following Boroson (1981) and using the marking the disk method.

The main results are given in Table 1, and presented in the different panels of Figs. 2 and 3. (NGC 718 data appears as a discrepant point in all the relations involving its B-magnitude. We suspect that the abnormally red colors we have measured are not correct and the galaxy should be observed again before being included in the discussion. This is the reason to omit it in the following.) No trend is found between the disk and bulge parameters and the morphological type. In particular, for the Scs in our sample it is clear that their disk and bulge properties span a big range. Indeed, the size of our sample is too small to draw conclusions. But the trend we find for the isolated galaxies is much alike to that shown by larger samples of non-interacting galaxies. We have to insist, before starting comparisons between different sets of data, on the differences that can be induced by the use of different methodologies. Thus, it has been argued that exponential rather than fits would be more appropriate for the bulges of late spirals (Andreadakis & Sanders 1994; de Jong 1996a; Courteau, de Jong & Broeils 1996, Seigar & James 1998). The resulting bulges are then fainter than when a law is fitted. Moreover, the use of 1-D bulge profiles (our case) produce bulges with fainter central surface brightness and larger effective radii than 2-D fits.

Fig. 2a-e. The bulge and disk properties of the isolated galaxies in the B band. versus is given in a . Panel b is the Kormendy relation for the bulges of isolated spirals. The bulge versus disk surface-brightness, scale length and luminosity are shown in panel c -e , respectively. In the figures we have also plotted schematically the regions covered by the data obtained by de Jong (1996b) for non-interacting, non-perturbed spirals, represented by closed boxes.

Fig. 3. Relations for bulge and disk parameters. In the left panels we plot data from all spirals with type I profiles from Baggett et al. (1998) (196 galaxies). Sa to Sb are plotted as open triangles, Sbc to Scd as open squares and later than Scd as open circles. In the right panels we only plot the galaxies selected from Baggett et al. as isolated (filled triangles), together with our sample spirals (dark circles).

Fig. 3. (continued)

With all this in mind, we can compare our results with the B-band data presented by de Jong (1996b) for a sample of non-perturbed, non-peculiar spiral galaxies, a sample that, as we already argued, can be taken as not too dissimilar to ours except in the luminosity range. In the same Fig. 2 we also present de Jong's data. It is clear that there is a large overlap between both sets of results.

The slight differences that can be appreciated after a more detailed look, can be explained in terms of the differences in methodology and the already quoted bias towards luminous galaxies in our sample. Thus, the mean value of the central disk surface brightness that we find is = 20.9 0.6, rather on the bright end of the values determined by Bosma & Freeman (1993) and by Giovanelli et al. (1994), and higher than the value given by de Jong. The same is found for the derived disk luminosities. This is certainly due to the absence of galaxies fainter than M_B = -18 in our sample since, as pointed out by de Jong (1996b) and Courteau (1996), the average value of the central surface brightness of the disks (Freeman 1970) depends on the luminosity range considered. Similar considerations apply to the bulge parameters we have derived.

We find that the scale length values for the disks depends on the photometric band (see Table 1), in the sense that it becomes smaller for redder wavelengths. Evans (1994) has argued that this would be due to the effect of dust layers in non-transparent disks, but de Jong (1996c), who found a similar result to ours, was able to model this behaviour by combining the presence of disk gradients in both, stellar age and metallicity.

On the other hand, the relations between the surface brightness and scale parameters for both components appear to be better correlated for our galaxies than for de Jong's data (see panels a - disk - and b - bulge - of Fig. 2). The Kormendy relation for the bulge is significantly tighter for our data (in fact, de Jong reported no correlation between and for his data). A similar relation appears for the disk, with a scatter much lower than previously found.

The relation between the corresponding parameters of the bulge and disk components are presented in the panels c, d and e in Fig. 2. It can also be seen that the range of values spanned by the disks is significantly smaller than that of the bulges. This would indicate that the spiral galaxies differ between them mainly by the bulge properties, the disks being much more similar.

We have also compared our results with the data presented by Baggett et al. (1998) in the V band (Fig. 3). The comparison indicates the same trends already noticed: The relations between parameters become tighter when only isolated or similar galaxies are considered. And, as before, the disk parameters span a significantly smaller range than the bulge parameters.

Thus, the trends shown by our data, in spite of the small sample we have, are supported as physically meaningful when larger samples of galaxies in acceptably similar conditions are considered. The data indicates that for isolated or non-interacting galaxies there is a tight relation between the surface brightness and size not only for the bulge (Kormendy relation) but also for the disk, and that these relations are more scattered when interacting galaxies are added. We will discuss in Paper II whether this is related to the lack of faint galaxies in our sample or to the interaction status. On the other hand, the disks of different spirals, no matter their morphological types, are much more alike than their bulges.

4.2. The kinematical properties. The masses and M/L ratios

The parameters describing the rotation curves are given in Table 10 of Paper I, whereas the median values are given in Table 1 here. We recall that the inner gradient, G (in km s^-1kpc^-1) is defined as G = (/sin(i))/, where (in kpc) is the radius of the inner region of solid-body rotation, (in km s^-1) is the observed velocity amplitude at , and i is the disk inclination, as derived from the optical images. Other parameters in the table are (in kpc), the radius at the point of maximum rotation velocity, = /sin(i); R_M, the radius at the last measured point in the rotation curve, with its corresponding velocity. The mass has been evaluated at , for a simple homogeneous and spherically symmetrical distribution, M₂₅ = 2.3265 10⁵ M (Burstein & Rubin 1985).

Our results for V_M and M₂₅ are in agreement with those found for Sb, Sbc and Sc galaxies by Rubin et al. (1982). With the mass calculated as described, we have evaluated the M/L ratio for different bands. The median total mass-to-luminosity ratio is = 4.1 1.3, spanning a range from 1.7 to 7.6. For the other bands we find similar ranges and central values, 5.5 2.0 in V, and 4.5 2.7 in I.

We choose the parameter to describe the overall shape of the rotation curve farther than . In other words, the velocities are normalised to the maximum amplitude and the radii to . The parameter can take values around zero (for flat rotation curves), positive (for rising rotation curves), or negative (for declining rotation curves). The median value obtained for isolated galaxies, 8 , is compatible with flat rotation curves.

4.3. The star formation properties

A number of emission line regions, including all the nuclei we have observed, are present in our long slit spectra. Due to the wavelength coverage only the [OI]6300, H, [NII]6548,6583 and [SII]6713,6731 lines could be detected. Our analysis will mainly concentrate on the properties of the strongest H and [NII]6583 lines.

The general aspect of all the spectra of the detected emission line regions, including the nuclei, is that of normal HII regions photoionized by stars. The [SII]6713,6731 line ratio is always 1, indicating low electronic densities, as expected for such regions.

The median value of EW(H) is 11 Å and 18 Å for the nuclei and the external regions, respectively. The [NII]6583/H line ratio ranges from 0.12 to 0.8 for all the nuclei with EW(H)2 Å, with a median value of 0.49. Indeed, for the 3 nuclei with EW(H)2 Å, the line ratio is very high, but this is due to the fact that the measure of the H line intensity is severely affected by the underlying absorption.

For 104 of the 105 non-nuclear HII regions detected in our major axis long slit spectra the [NII]6583/H line ratio could be measured. The median value is 0.38, with 75 of the cases between 0.30 and 0.50. There are 4 extreme cases, with very high values, that are due to the presence of relatively strong absorption under H.

Thus the central values for nuclear and external HII regions are rather similar, with a larger range for the external regions. These results will be considered in more detail in Paper III, where they will be compared with data for HII regions in interacting galaxies.

Concerning the results from H CCD photometry, we measured 7 isolated galaxies, whose total H fluxes are given in Table 7 of Paper I (we have excluded NGC 718 from this analysis, since our spectroscopic data show that the contamination from [NII] emission lines is very important, see Paper I). The median value for the total H luminosity is log(H) = 7.89 0.55 (in solar luminosities), that is well within the range found for spirals (Young et al. 1996, and references therein). The H luminosities are well correlated with both the optical area of the galaxies and their FIR luminosities. From the first relation we derive H. This would mean than the average star formation per unit area, as judged from the H luminosity is about the same in all isolated spiral galaxies.

The FIR luminosity was calculated as in Young et al., , where D is the distance to the galaxy in Mpc, C is a correction factor for the flux longwards of 120 µm and shortwards of 40 µm (they depend on / and are taken from Table B.1 of the Catalogued Galaxies in the IRAS Survey, Londsdale et al. 1985) and and are the IRAS 60 and 100 µm fluxes in Jy. In Fig. 4 we plot the comparison of H and FIR luminosities for our sample galaxies, together with the points corresponding to the 10 galaxies catalogued as isolated by Young et al. Both samples of isolated spirals agree within the errors. They are all in the region occupied by normal spirals, compatible with ionization produced by O5 to B0 stars. We anticipate here that this is not the case for all the spiral in isolated pairs (Paper III).

Fig. 4. Relationship between integrated H and FIR luminosities. Solid dots are for our isolated spiral galaxies, triangles are for the sample of isolated galaxies by Young et al. (1996). The solid lines represent the expected relations for O5, O9 and B 0 stars, as presented by Devereux & Young (1990)

Finally, we have also analysed the star formation history of those galaxies for which we have a complete set of data, i.e., the B-luminosity, the H integrated flux, and the total mass. Following Gallagher et al. (1984), we quantify the present star formation rate in terms of the H luminosity, the star formation rate during the past 10⁹ years as a function of the B luminosity, and the initial star formation rate as a function of the total mass. Unfortunately we have all the relevant data for only 5 galaxies in our sample. Still, the result is compatible with a constant star formation rate along their lifetimes, with a smooth SFR, for the five galaxies. We will report in Paper III that this is not the case for some of the spiral galaxies in isolated pairs.

© European Southern Observatory (ESO) 1999

Online publication: March 18, 1999
helpdesk.link@springer.de