5 Conclusions

The structures of planetary nebulae are difficult to determine with any degree of reliability - and this, indeed, probably represents one of the primary obstacles to a full comprehension of their optical spectra. This apart, however, it is clear that spectra in most planetary nebulae can be simulated tolerably well through the application of comparatively simple models. Thus, in the present work, we have shown that [OII], HeI, and HeII intensities are all more or less consistent with what would be expected given radiative excitation, a spherical nebular structure, and a broad range of central star temperatures, gas densities and (less critically) abundances.

Whilst higher excitation lines appear to possess relative strengths which are invariant with nebular radius, [OI], [SII], and [NI] are all shown to display evolutionary changes, together with an appreciable scatter which is likely to derive from variations in T_*, $n\rm _e$, and the spectral sampling locations. The absence of comparable scatter in [OIII] and HeI (and the relative invariance of these line ratios) arises from a comparability of emission zones to that for HI, and relative insensitivity of emission coefficients to temperature and density.

In contrast to the other transitions, it appears that [OI] intensities exceed model predictions in the majority of nebulae. Such emission may arise through a variety of radiative processes associated (perhaps) with neutral shell components, including UV shadowing, charge-exchange reactions and so forth. In addition, we note that kinematic expansion velocities may be sufficient to trigger appreciable shock excitation of [OI]; a process which would explain the close correlation between [OI] and H₂ S(1) line intensities.

Comparison between observed line ratios, and radiative and shock modelling trends also suggests that a suprisingly large proportion of [SII] $\lambda 6716/31$ Å (approximately $\sim 0.5$ on average) may arise through shock excitation. This, of confirmed, would suggest that [SII] density estimates may be biassed towards the higher values characterising post-shock emission zones.

We have identified 14 nebulae in which shock emission may be appreciable, certain of which (CRL 618, M2-56) have previously been identified as shock sources on the basis of spectral diagnostics. A large proportion of these candidates appear also to be characterised by appreciable shocked H₂ S(1) emission. It is suggested that differences in spectral line ratios between bipolar nebulae and the more general nebular sample may also derive from shocks; for which case, it is apparent that most BPN are likely to be shocked, and mean shock velocities would be of order $V\rm _s = 80\Rightarrow 100~km~s^{-1}$. Such values appear consistent with observed BPN wind outflow velocities, implying that much of the observed excess may derive from interaction between high velocity winds and the very much slower primary shell outflows.

Sources containing FLIERs, on the other hand, appear to be located at the opposite end of the excitation scale, and are confined to extremely tightly defined spectral regimes; of the spectral measurements having log(10²[OII]/H$\beta$) $\leq$ 1.4 and log(10²[SII]/H$\beta$) $\leq$0.4, approximately 52% correspond to sources associated with FLIERs. This characteristic suggests a new diagnostic for the identification of further such outflows; a methodology which has been applied to suggest possible FLIER activity in NGC 6537, NGC 6879, IC 351 and (in particular) J320, where various symmetrically disposed pairs of condensations straddle the nucleus and dominate the low excitation structure.