next previous
Up: Spatially resolved spectroscopy of


   
3 Stellar kinematics

We measured the stellar kinematics from the galaxy absorption features present in the wavelength range and centered on the Mg line triplet ( $\lambda\lambda~5164,~5173,~5184$ Å) by applying the Fourier Correlation Quotient method (Bender 1990) as done by BSG94.

The spectra were rebinned along the dispersion direction to a natural logarithmic scale, and along the spatial direction to obtain a nearly constant $S/N \geq 20$ per resolution element. In a few spectra the S/N decreases to $\sim$10 at the outermost radii. The galaxy continuum was removed row-by-row by fitting a fourth to sixth order polynomial. The quality of the final spectrum depends on the resulting S/N. In Fig. 1 we show examples of central spectra covering the 3 quality classes listed in Table 1. The quality parameter is 1 for $S/N\geq100$, 2 for  $50 \leq S/N < 100$, 3 for 20 < S/N < 50.


  \begin{figure}
\par\includegraphics[width=8.8cm,clip]{MS2880f1.eps}\end{figure} Figure 1: Examples of central spectra covering the range of quality classes. Relative fluxes have false zero points for viewing convenience.


  \begin{figure}
\par\includegraphics[width=8.3cm,clip]{MS2880f2a.eps}\hspace*{1cm...
....eps}\hspace*{1cm}
\includegraphics[width=8.3cm,clip]{MS2880f2d.eps}\end{figure} Figure 2: Kinematical parameters and line indices measured along the observed axes of the sample galaxies. For each axis the curves are folded around the nucleus and filled dots and asterisks refer to the two sides of the galaxy. The radial profiles of the line-of-sight velocity (v) after the subtraction of systemic velocity, velocity dispersion ($\sigma $), third (H3) and fourth (H4) order coefficient of the Gauss-Hermite decomposition of the LOSVD are shown in the left panels (from top to bottom). The radial profiles of the line indices H$\beta $, [MgFe], <Fe>, Mgb and Mg2 are plotted in the right panels (from top to bottom).


 \begin{figure}
\par\includegraphics[width=8.3cm,clip]{MS2880f2e.eps}\hspace*{1cm...
...ps}\hspace*{1cm}
\includegraphics[width=8.3cm,clip]{MS2880f2h.eps}
\end{figure} Figure 2: continued.


 \begin{figure}
\par\includegraphics[width=7.5cm,clip]{MS2880f2i.eps}\hspace*{1cm...
...ps}\hspace*{1cm}
\includegraphics[width=7.5cm,clip]{MS2880f2l.eps}
\end{figure} Figure 2: continued.


  \begin{figure}
\par\includegraphics[width=7cm,clip]{MS2880f3a.eps}\hspace*{1cm}
\includegraphics[width=7cm,clip]{MS2880f3b.eps}\end{figure} Figure 3: Kinematical parameters (v, $\sigma $, H3, and H4) and line indices profiles ($H\beta $, [MgFe], <Fe>, Mgb and Mg2) along the major axis of GMP 1176 (left panels) and GMP 3792 (right panels) measured during run 1 (open symbols) and run 4 (asterisks) compared to those obtained by Mehlert et al. (2000, filled symbols). For each run different symbols refer to the different sides of the galaxy.

To measure the stellar kinematics of the sample galaxies we adopted HR 6817 (K1III) as kinematical template for runs 1, 2 and 3 and HR 3427 (K0III) for run 4 and we considered the wavelength range 5115-5541 Å (see Fig. 1) around the Mg lines of the galaxies. We derived for each galaxy spectrum the line-of-sight velocity distribution (LOSVD) along the slit and measured its moments, namely the radial velocity v, the velocity dispersion $\sigma $ and the values of the coefficients H3 and H4. At each radius, they have been derived by fitting the LOSVD with a Gaussian plus third- and fourth-order Gauss-Hermite polynomials ${\cal H}_3$ and ${\cal H}_4$, which describe the asymmetric and symmetric deviations of the LOSVD from a pure Gaussian profile (van der Marel & Franx 1993; Gerhard 1993). Errors on the LOSVD moments were derived from photon statistics and CCD read-out noise, calibrating them by Monte Carlo simulations as done by BSG94. In general, errors are in the range of 3-10  $\rm km~s^{-1}$ for v and $\sigma $, and of 0.01-0.04 for H3 and H4, becoming larger in the outer parts of some galaxies where for $10 \leq S/N < 20$. These errors do not take into account possible systematic effects due to template mismatch or the presence of dust and/or faint emission. The measured stellar kinematics are reported in Table 3 and plotted in Fig. 2.

Figure 3 shows the comparison between the measurements of v, $\sigma $, H3, and H4 along the major axis of GMP 1176 and GMP 3792 obtained here and measurements obtained in Paper I. The value derived from the different datasets are in agreement within the errors. However in the case of GMP 3792 the velocity dispersion measured in run 4 is systematically lower than that measured both in Paper I and in run 1.


next previous
Up: Spatially resolved spectroscopy of

Copyright ESO 2002