Existence & parameters of thick disks in late-type disk galaxies

In seven of the ten galaxies we fitted we find a thick disk. Two galaxies show an extended vertical structure that cannot be described by a two component fitting function. One galaxy does not show a clear sign of a thick disk component, although the two component fitting function is able to fit the profiles. The results give thick disks in the range of $\mu_{n-k}$ $\approx$ 3.4-5.6 mag arcsec$^{-2}$, $f_z$ $\approx$ 3.3-7.7 and $f_h$ $\approx$ 0.6-3.4.
Determining the quality of the fits to the thick disk components in late-type disk galaxies is complicated as the profiles are usually not smooth and the quality is different for each cut. Tests on artificial galaxies (see Section 4.6.3), which have perfect profiles for the outer component showed that the fitting method is often not capable of producing all the correct parameters as were set as input parameters. As we have to find the parameters empirically, it is hard to determine which kind of deviation our results are subject to, especially as the profiles of the sample galaxies have a much lower quality than the artificial galaxies. Because of this we mention here average limits within one can expect to find the thick disk parameters and to not hold unto all the individual results as they can have unknown deviations.
We also noted the possible behaviour of the fit function to couple $\mu_{n-k}$ and $f_h$ to improve the fit, but it is not clear when this happens as the profiles of each galaxy are different. With six parameters to fit the profiles there exists a large range of possible solutions for the best fit. However, we showed in our tests on artificial galaxies that our fitting method is able to reproduce the input parameters sufficiently where it concerns the thick disk component in case the real galaxy would follow the model.
In Table 11 three galaxies give a $\chi^{2}_{\nu,1}/\chi^{2}_{\nu,2}$ ratio higher than 10, presenting very certain thick disk cases. Three galaxies are below the break point and their fits can thus be considered to be a bad representation of a possible thick disk component. The other galaxies have ratios just above the break point, showing that there is a strong hint for a thick disk, but no absolute certainty as there still exists a risk of being biased by residual background structure causing the thick disk scaleheigth to be overestimated.
Although we are able to fit thick disks in our sample galaxy, most of regions dominated by the thick disk lie under $\mu_\textrm{\scriptsize crit}$. This is to be expected as it is faint and hard to detect. Several galaxies show the characteristic bending of a second component just above or around $\mu_\textrm{\scriptsize crit}$, which supports the claim for a thick disk component in those galaxies.
The relative scaleheight of the thick disk shows a possible relationship to a vertically extended size of the bulge, which can seen from the contour maps of the galaxies. The high values for $f_z$ for NGC 2424, NGC 5290 and UGC 4277 can be related to those, while the low values of IC 3322A and NGC 5981 can be matched to their small bulges. This might suggest the not seen thick disk component of NGC 1886 is rather shallow as is has a strong bulge component.
Overall, the thick disk parameters all point at faint thick disks and higher scaleheights for late-type disk galaxies, as previous results gave $\mu_{n-k}$ of 4 mag arcsec$^{-2}$ and $f_z$ values no higher than 6.5 (see Section 5.4).