Faint Haloes Of Planetary Nebulae

Planetary nebulae (PNe) indicate that the last phases of stellar evolution of low to intermediate mass stars (up to 8 solar masses), when the star has consumed all of its fuel for nucleosynthesis, move to a position in the HRD diagram on the tip of the Asymptotic Giant Branch (AGB), where it is extremely luminous, yet has a low surface temperature and looses copious amount of matter through strong dust-driven stellar winds. This mass-loss phase is of considerable interest for our quantitative understanding of the cycle of matter in galaxies, and the enrichment of the interstellar medium. However, the theory of mass-loss on the AGB is complicated, involving chemical processes of production and destruction of dust, stellar pulsations, and so forth. To date, different treatments of these problems have only provided largely discrepant predictions, which is why observational tests are badly needed. The PN phase, however, which follows the final period at the tip of the AGB, offers a unique diagnostic opportunity. The nebula forms when the remainder of the star shrinks and heats to temperatures of between 104 to 105 K and thus ionizes the previously ejected matter. Measuring the temperature, density, and chemical composition as a function of radius throughout the nebula, would, in principle, allow one to study the different shells as fossil records of the mass-loss history over long periods of time. While the inner parts of PNe are usually bright in a rich spectrum of emission lines and well accessible for plasmadiagnostic tools. The most interesting early phases of mass-loss are recorded in the extremely faint haloes of PN, which have typically surface bright-nesses on the order of

10- to 10- erg/cm/sec/arcsec or fainter, even for the most prominent lines like e.g. Ha, or [O III] 5007 A. For the same reasons as explained in Sec. 2, we have employed the PMAS LARR IFU and the N&S mode of observation to increase sensitivity and sky subtraction accuracy for this challenging task, which would otherwise require substantial dark time at 8-10 m class telescopes with conventional slit spectrographs for any hope for success. Figure 6 illustrates, for an extreme case, the contrast between the faint halo emission lines (bottom spectrum), which measure on the order of 1/100 of the sky continuum, which essentially is a solar spectrum with a wealth of absorption lines (top), e.g. the prominent Ca H+K feature, and many others.

These spectra are extreme in the sense that the observations were obtained 4 days from a full moon, i.e. very strongly background-limited. However, the example demonstrates the power of binning and N&S spectroscopy. As shown in Figure 7, the blue spectra obtained from the halo region outlined in the insert of Figure 8 (reconstructed map in [O II] 3727/3729 A), where the PMAS guiding camera frame (narrowband Ha) shows no trace of emission, are just presenting noise, if only a single spaxel is selected (top panel).

Figure 6. Sky spectrum in the blue (top), object (bottom).
Figure 7. Signal-to-noise gain with binning of spatial pixels. (top) Single spaxel, (middle) 9 spaxels, (bottom) 256 spaxels. The gains are 3, and 16, respectively.

Figure 8. Halo of NGC6720, observed with PMAS IFU.

Figure 8. Halo of NGC6720, observed with PMAS IFU.

Coadding 3^3 spaxels in the same region immediately improves the SNR by a factor of 3. When finally collapsing the entire IFU into a single spectrum, a SNR gain of 16* is achieved, now revealing the faint emission lines of [O II], [Ne III], Hy, etc. Note that the final spectrum exhibits a r.m.s. of 2*10-18 erg/cm2/sec/arcsec2, which is very faint considering that the spectrum was obtained with a total on-target exposure time of 3600 s near full moon. Note also that the data were obtained in N&S mode, which proved crucial for obtaining the required sky subtraction accuracy to significantly better than 1%. We have obtained comparison sky spectra without N&S, showing that the sky varied in intensity and colour on the order of several percent within less than 1800 s. A peculiar problem presented the presence of terrestrial emission lines, e.g. Hg 4358 A, which completely jeopardized any attempt to measure the important nearby, but faint, diagnostic [O III] 4363 A line that was completely swamped by a strong Hg line residual when subtraction the comparison sky exposure in the conventional way. However, as can be appreciated from the bottom panel of Figure 7, the spectrum resulting from N&S observing is completely consistent with random background noise, without any apparent sky line residuals. Although we were unable to recover the [O III] 4363 A line owing to its faintness and the continuum sky noise floor near full moon, we estimate that with 2-3 hours of observing under grey to dark conditions the line will become measurable. In addition to our example of NGC6720, we have obtained data from 3 observing runs in Feb. 2003 and Aug. 2004 for a total of 6 PNe, with more observations expected in September 2005.

Was this article helpful?

0 0
Telescopes Mastery

Telescopes Mastery

Through this ebook, you are going to learn what you will need to know all about the telescopes that can provide a fun and rewarding hobby for you and your family!

Get My Free Ebook

Post a comment