SpringerLink
Forum Springer Astron. Astrophys.
Forum Whats New Search Orders


Astron. Astrophys. 355, 713-719 (2000)

Previous Section Next Section Title Page Table of Contents

3. Data analysis

[O III] and H[FORMULA]+[N II] emission-line objects were selected in our images in the following way. We first computed the scale factors to be multiplied to the continuum images in order to subtract the stellar background of M33 from the [O III] and H[FORMULA]+[N II] frames. That was done by dividing the fluxes in the continuum of a large number of stars of M33 by those in the emission-line filters. The stars were automatically selected using DAOPHOT. The scaled continuum image was then subtracted from the [O III] and H[FORMULA]+[N II] images. The scale factor resulted to be around 0.7 for the correction of the continuum to both the [O III] and H[FORMULA]+[N II] frames. This procedure removed the stellar background and isolated emission-line objects, essentially HII regions and PNe. Note that the continuum filter we used has a central wavelength which is relatively far from that of the emission-line filters (especially for the H[FORMULA]+[N II] one). As a consequence, a small fraction of stars, with spectral energy distributions significantly different from "typical" stars of M33 used to compute the scale factors, are not perfectly subtracted. However, since they usually appear as "holes" either in the [O III] or H[FORMULA]+[N II] subtracted image, they can be easily distinguished from genuine emission-line objects.

The typical size of a galactic PN is of some tenths of a parsec (cf. Peimbert 1990), while evolved PNe have been found with sizes of up to 4 pc (e.g. Corradi et al. 1997). At the distance of M33 (840 kpc, Freedman et al. 1991), 1" corresponds to 4 pc and therefore we do not expect to resolve any PN.

For this reason, any extended emission region in our images was considered to be an HII region or a SN remnant. Candidate PNe were consequently selected among emission-line objects which are not spatially resolved. We found 197 objects of this kind, i.e. which have positive emission in the [O III] and/or in the H[FORMULA]+[N II] subtracted images, and with a spatial FWHM within 3[FORMULA] of the mean value computed using many field stars. Among them, 63 sources have a non-negligible continuum emission, i.e. the signal in the continuum filter at the position of the object is enhanced compared to its surroundings. Negligible continuum is expected from a PN, unless the central star is very luminous or has a bright companion or a foreground/background star of M33 projects at the same position as the PN. Conservatively we have excluded these 63 objects and considered as bona fide candidate PNe the remaining 134 objects listed in Table 2.


[TABLE]

Table 2. PN candidates found in M33. Fluxes in H[FORMULA]+[N II] and in [O III] are in units of 10-15 erg cm- 2 s-1. [O III] magnitudes and their adopted, individual errors, are indicated in the last two columns.



[TABLE]

Table 2. (continued)


[O III] and H[FORMULA]+[N II] line fluxes were measured using APPHOT in the subtracted images, with photometric apertures three times the FWHMs. Errors in the fluxes were estimated considering both photometric errors, given by Poissonian statistic on background and source photons and by the detector noise, and the scattering between fluxes of the same objects measured in two different images. Photometric errors vary between few percents for the brightest objects to about 20[FORMULA] for the fainter ones, while systematic errors are of about 5[FORMULA] for [O III] and about 15[FORMULA] for H[FORMULA]+[N II].

An astrometric solution was computed for each frame, using the positions of bright stars in M33 given in the National Geographic Society-Palomar Observatory Sky survey. The accuracy of the coordinates of the 134 candidate PNe, which are listed in Table 2, is of about 1". Their location within M33 is indicated by +'s in Fig. 2.

In Table 3, we list the 63 compact emission-line objects with non-negligible continuum, together with their measured [O III] and H[FORMULA]+[N II] fluxes. Since no absolute calibration for the Strömgren Y filter was available, the contribution of the continuum relative to [O III] was estimated by measuring the intensities of sources in the scaled Strömgren Y frames. With the present data, it is difficult to say whether these emission-line object are compact HII regions, Be stars, symbiotic stars, Wolf-Rayet stars, LBVs, etc. Further spectroscopic studies are needed to clarify their nature.


[TABLE]

Table 3. Emission lines and continuum objects found in M33. H[FORMULA]+[N II] and [O III] fluxes are in units of 10[FORMULA]. The last column gives the fluxes measured in the scaled Strömgren Y image, relative to the [O III] ones (see text).



[TABLE]

Table 3. (continued)


Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: March 9, 2000
helpdesk.link@springer.de