next previous
Up: High-velocity interstellar clouds towards


Subsections

2 Observations and data reduction


2.1 The sample

The programme stars were chosen from the UBV photometry of Buonanno et al. (1983) and Battistini et al. (1985), using the following selection criteria:
  
Table 1: Sample of stars in the nearby field of M 15 used to search for HVC absorption

\begin{tabular}
{lcccc}
\hline 
Star & $V$&$ B-V$& $ l $& $b$ \\  & (mag) & (mag...
 ...
\\ \multicolumn{5}{l}{{\small $b$: Previously unidentified star.}}\end{tabular}

  
\begin{figure}
\vspace*{1mm}
\includegraphics[height=9.5cm,width=6.5cm]{7947.f1.eps}\vspace*{-2mm}\end{figure} Figure 1: Programme stars observed in the globular cluster M 15. Stars within the region enclosed by the dotted lines have their equivalent widths and radial velocities plotted in Fig. 4
Table 1 lists the positions and magnitudes of stars that were found to meet these criteria, and their positions are plotted in Fig. 1. The star identifications in Table 1 correspond to those in Table II of Buonanno et al. (1983) and Battistini et al. (1985). Observations of K-127 were serendipitously obtained with K-080, and N-1 with K-143 and K-154.


2.2 ISIS observations of cluster stars


The observations were carried out using the ISIS spectrograph on the 4.2 m William Herschel Telescope at La Palma Observatory on 16/17 August 1997. The blue and red arms were operated with the H2400B lines/mm grating (Loral $2048 \times 2048$ CCD) and R1200R lines/mm grating (Tek $1024 \times 1024$ CCD), respectively. The spectra were centered on 3960 Å  for the Caii K and H lines (blue arm) and 5900 Å  for the Nai D lines (red arm). This instrumental configuration resulted in a FWHM velocity resolution of $\sim 42\,{\rm km\,s^{-1}}$ for the Nai data and $\sim 30\,{\rm km\,s^{-1}}$ for the Caii data, determined from CuAr arc lines used for wavelength calibration. The slit width was $1\hbox{$.\!\!^{\prime\prime}$}03$, with a scale of $0\hbox{$.\!\!^{\prime\prime}$}21$ per pixel for the blue arm and $0\hbox{$.\!\!^{\prime\prime}$}33$ per pixel for the red arm. Multiple exposures were made for each star, interspersed with CuAr arc line calibration exposures. Bias and flat field exposures were made at the start and the end of each night.


2.3 Data reduction


The raw CCD images were reduced using standard techniques within the NOAO IRAF package (Morris & Privett 1995). The spectra were wavelength calibrated, and cross-correlation techniques were employed for multiple spectra of the same object to ensure that there was no shift between successive exposures. Images were then co-added, using weights determined by their continuum signal-to-noise levels. Typically, co-added spectra have a signal-to-noise ratio (S/N) of about 50 for the red data, and in the range 7 - 20 for the blue data. The spectra were then shifted to the Dynamical LSR (Mihalas & Binney 1981), using corrections obtained from the STARLINK RV (Wallace & Clayton 1996) package.

Although the Caii K and H lines are free of telluric absorption features, the Nai D lines show evidence of contamination. However, a comparison of a synthetic telluric spectrum with the observational data demonstrated that only the continuum level was affected, and that the telluric lines would not affect the measurement of HVC radial velocities. Therefore we did not remove these telluric lines from the data.

  
Table 2: Equivalent width and radial velocity data for both the stellar and interstellar Caii K lines

\begin{tabular}
{lcccccccc}
\hline 
\multicolumn{2}{l}{Sightline} & \multicolumn...
 ...interstellar gas.
Uncertainties are fitting errors at $1 \sigma$.}}\end{tabular}


  
Table 3: Radial velocity data for the interstellar Nai D lines

\begin{tabular}
{lccc}
\hline 
\\ Star &$v_{\rm LIS}$\space & $v_{\rm HVC}$ & $v...
 ...umn{4}{l}{{\small Uncertainties are fitting errors at $1 \sigma$.}}\end{tabular}



next previous
Up: High-velocity interstellar clouds towards

Copyright The European Southern Observatory (ESO)