2. Observations and analysis
CCD narrowband images of 51 symbiotic stars were obtained between 1991 and 1997 during four observing runs at the European Southern Observatory (ESO), La Silla, Chile, and at the Observatorio del Roque de los Muchachos (ORM) on La Palma, Spain. The telescope used, date of observation, CCD type and pixel size are listed in Table 1.
Table 1. Observing campaigns
When searching for extended ionized nebulae around symbiotic stars, the main problem is to avoid all the disturbing effects (saturation, instrumental internal reflections, extended wings of the point-spread function, etc.) produced by the strong emission from the unresolved central source (in the visible, mainly the cool giant and its ionized wind). Because of that, faint or small nebulae can easily escape detection because they are masked by the noise of the emission wings of the central source. For this reason using a filter centred on , which is a typical choice to image ionized nebulae, is generally not a good solution in the case of symbiotic stars, since the hydrogen line emission from their cores is extremely strong. In our experience, the best choice is to use a narrow filter (FWHM2.5 nm) centred on the [NII ] line at =658.3 nm. The reason is that the [NII ] emission is generally fainter than H in the central source, but strongly enhanced in the extended nebulae.
In particular, [NII ]658.3 is the strongest optical line in most extended nebulae around symbiotics (e.g. Schwarz 1991, Corradi & Schwarz 1993, Corradi et al. 1999). The same occurs in related objects, such as bipolar planetary nebulae, where the [NII ]658.3/H ratio is usually larger than unity and can be as high as ten (Perinotto & Corradi 1998). As an example, the large () nebula around HM Sge escaped detection in images taken by us in 1991 using a relatively broad filter including both the H and [NII ] emission, while it was clearly revealed in our 1996 images obtained through a narrow [NII ] filter (Corradi & Schwarz 1997). An alternative to [NII ] is to use a filter centred on the near-UV [OII ] line at 372.8 nm, with the advantage that the emission from the cool giant is several magnitudes fainter than in the red, but with the disadvantage that the nebular [OII ] emission is also fainter than [NII ]. Using filters in low excitation lines as [NII ] or [OII ] prevents detection of high excitation material, but this is usually confined to the innermost regions of the symbiotic nebulae. With very narrow filters, such as the 1.0 nm wide [NII ] filter used in our campaigns at the NOT, high velocity outflows (300 km s-1 or even less depending on the systemic velocity of the object) can also be missed, because the line is Doppler shifted out of the pass band of the filter. In our survey, we used a variety of filters depending on the choice available at the time of the observations (see Table 5).
To reveal small or very faint nebulae, excellent seeing conditions are also very important. The use of a proper coronograph also helps. In our observing run at the NOT in 1997, in some cases we used a coronographic spot placed in the focal plane of the telescope to stop the light from the central stars, which allowed us to take deeper exposures avoiding charge blooming due to oversaturation of central stars, but which did not eliminate the instrumental artifacts in the wings of the point-spread function (such as the diffraction spike image of the secondary mirror spider).
The detection limit in our survey varied according to the telescope, exposure time, pixel size, and relative strength of the "disturbing" emission from the central source; under the best conditions, it was 10-16 erg s- 1 cm-2 arcsec-2 in the [NII]658.3 line.
Apart from the frames in which an extended nebula was obviously detected, in all the other frames we compared the full width at half maximum (FWHM) of the image of the symbiotic star with that of the field stars to check for the presence of nebulae only slightly larger than the seeing value. One limitation to this analysis is that, with some of the imagers that we used, the point-spread function varies throughout the field. For this reason, only variations larger than 10 and which are systematic in different images were considered to be significant for the detection of a resolved nebula. In these cases, an off-band image was also taken to check whether the extended region is an emission-line nebula. The results are reported in Tables 2 to 5, where symbiotic systems are listed in order of increasing right ascension.
© European Southern Observatory (ESO) 1999
Online publication: March 1, 1999