next previous
Up: Be stars in and Clouds


3 Selection of Be and other H$\alpha$ emitting stars

  In order to detect stars with H$\alpha$ emission, plots were made of the colour index $R-{\rm H}\alpha$ against V-I for all the stars which were common to the V, I, R and H$\alpha$ frames. Figures 1 (NGC 330), 2 (NGC 346), 3 (NGC 1818), 4 (NGC 1948), 5 (NGC 2004) and 6 (NGC 2100) show the ($R-{\rm H}\alpha$, V-I) diagrams produced. As noted in Sect. 2, the $R-{\rm H}\alpha$ zero-point is arbitrary in these plots, although the zero-point was chosen so that normal main-sequence stars had $R-{\rm H}\alpha\approx 0$.

  
\begin{figure}
\resizebox {\hsize}{!}{\includegraphics{ds7804f1.eps}}\end{figure} Figure 1: The (V, V-I) and ($R-{\rm H}\alpha$, V-I) diagrams for the cluster NGC 330 (panels a and c) and its surrounding field (panels b and d). The usable cluster area lies between radii $15\hbox{$^{\prime\prime}$}<r<1\hbox{$.\mkern-4mu^\prime$}6$ while the field is defined as $r\gt 1\hbox{$.\mkern-4mu^\prime$}6$. Be stars are identified as the stars above the dashed line in the ($R-{\rm H}\alpha$, V-I) diagrams and they are shown on the (V, V-I) diagrams as filled circles

  
\begin{figure}
\resizebox {\hsize}{!}{\includegraphics{ds7804f2.eps}}\end{figure} Figure 2: The same as Fig. 1, but for NGC 346. The usable cluster area lies between radii $15\hbox{$^{\prime\prime}$}<r<2\hbox{$.\mkern-4mu^\prime$}2$ while the field is defined as $r\gt 2\hbox{$.\mkern-4mu^\prime$}2$

  
\begin{figure}
\resizebox {\hsize}{!}{\includegraphics{ds7804f3.eps}}\end{figure} Figure 3: The same as Fig. 1, but for NGC 1818. The usable cluster area lies between radii $15\hbox{$^{\prime\prime}$}<r<2\hbox{$.\mkern-4mu^\prime$}2$while the field is defined as $r\gt 2\hbox{$.\mkern-4mu^\prime$}2$

  
\begin{figure}
\resizebox {\hsize}{!}{\includegraphics{ds7804f4.eps}}\end{figure} Figure 4: The same as Fig. 1, but for NGC 1948. The usable cluster area lies between radii $15\hbox{$^{\prime\prime}$}<r<2\hbox{$.\mkern-4mu^\prime$}7$ while the field is defined as $r\gt 2\hbox{$.\mkern-4mu^\prime$}7$

  
\begin{figure}
\resizebox {\hsize}{!}{\includegraphics{ds7804f5.eps}}\end{figure} Figure 5: The same as Fig. 1, but for NGC 2004. The usable cluster area lies between radii $15\hbox{$^{\prime\prime}$}<r<2\hbox{$.\mkern-4mu^\prime$}0$while the field is defined as $r\gt 2\hbox{$.\mkern-4mu^\prime$}0$

  
\begin{figure}
\resizebox {\hsize}{!}{\includegraphics{ds7804f6.eps}}\end{figure} Figure 6: The same as Fig. 1, but for NGC 2100. The usable cluster area lies between radii $15\hbox{$^{\prime\prime}$}<r<1\hbox{$.\mkern-4mu^\prime$}4$ while the field is defined as $r\gt 1\hbox{$.\mkern-4mu^\prime$}4$

The stars with strong H$\alpha$ emission were selected from the above figures. Using Fig. 1c as a typical case, we note that most main-sequence stars lie in a tight clump around $R-{\rm H}\alpha=0.0$ and V-I=-0.15, while nearly all cooler stars form an almost horizontal band. These two groups of stars do not have detectable H$\alpha$ emission. However, there is another group of stars extending up and slightly to the red of the main-sequence clump: these stars clearly show significant H$\alpha$ emission and they are Be stars. Occasional cooler (giant) stars also exhibit strong H$\alpha$emission.

The dispersion in $R-{\rm H}\alpha$ of normal MS stars about the centre of the clump is determined largely by the photometric errors in the narrow-band H$\alpha$ magnitude, and this uncertainty will depend on the magnitude of the star. We limit our Be star selection process to stars with V < 17.5, which corresponds approximately to an error in $R-{\rm H}\alpha$colour of $\sim$0.2 magnitudes. In order to select a sample of Be stars, we draw a line in each of the $R-{\rm H}\alpha$ diagrams parallel to the sequence of non-emission line stars and at a distance of $\sim$0.4 magnitudes above it, which gives a high probability that we exclude all members of the clump of normal main-sequence stars.

Our observations were made under seeing conditions of 1.5-2$^{\prime\prime}$, and as a consequence the central regions of the clusters suffer from crowding. We therefore exclude the innermost 15$^{\prime\prime}$ (in radius) of each cluster from the Be star search. On each image taken, we select a radius at which the stellar density drops to a value indistinguishable from that of the surrounding field. Stars within this radius are assumed to belong to the cluster, and those outside to the field. The radius selected for each cluster is given in the captions of Figs. 1 to 6.

The positions, magnitudes and colours of the Be stars selected as described above are listed in Tables 3 (NGC 330), Table 4 (NGC 346), Table 5 (NGC 1818), Table 6 (NGC 1948), Table 7 (NGC 2004) and Table 8 (NGC 2100). Tables 3 to 8 are available in electronic form from the CDS. The astronomical coordinates in these tables were derived from the secondary astrometric standards of Tucholke et al. (1996). Finding charts for the Be stars are given in Figs. 14 and 15 (NGC 330), 16 and 17 (NGC 346), 18 and 19 (NGC 1818), 20 (NGC 1948), 21 and 22 (NGC 2004), and 23 and 24 (NGC 2100).


next previous
Up: Be stars in and Clouds

Copyright The European Southern Observatory (ESO)