Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 355, 922-928 (2000)

Previous Section Next Section Title Page Table of Contents

3. Energetics of LINERs

3.1. Classification of LINERs

Recently, more than a dozen of LINERs have been studied with space facilities or large ground-based telescopes. Table 2 lists these sources with claims for their energetics, except for M 81 because of the debate mentioned in Sect. 1. Fig. 2a indicates the positions where these LINERs are located in the FIR-radio diagram of WR galaxies. It is very instructive to see that the LINERs with SB-supported claims or with the existence of nuclear starburst, in notation of SB in Table 2, have a similar distribution to WR galaxies, while the AGN-supported LINERs or those with the existence of AGN are distinctively located in the upper-left part of the FIR-radio diagram. The latter category of objects is found above the upper border (solid line in Fig. 2a) of our models for starburst events.

[FIGURE] Fig. 2. a Same correlation diagram as Fig. 1a, superposed by LINERs with claims of SB-supported or AGN-supported ones, selected from literature. Dotted lines cover the area where the majority of WR galaxies are located. Solid line illustrates the prediction of starburst event at age of 107 yr. The notation of L in the lower-right box denotes LINERs. The notation of (L+AGN+Starburst) for M 81 is a symbol of "composite" for this source, see the text for details.
b Same as Fig. 2a, but the superposed LINERs are unclassified ones by symbols of starred triangles, extracted from Véron Catalog, and some identified types of LINERs by symbols of inverse-starred triangles, selected from Alonso-Herrero et al. (1999). The symbols of hexa-stars are referred to the common ones between the two sets.


Table 2. LINERs with claims

Furthermore, we investigate two cases of LINERs that are considered to have a composite nature, NGC6240 and M 81. For NGC6240, Schulz et al. (1998) claimed, based on ROSAT data, that both AGN and starburst contribute in roughly equal proportion to the energetics of this galaxy. In our Fig. 2a, both NGC6240 and M 81 are approximately located on the border of our model prediction. Due to the enrichment of dust and the influence of cool dust component as discussed in Sect. 2, the actual position of our model prediction in Fig. 2a might somewhat move to the right and curve up at small burst strength.

Of particular importance to the AGN nature in LINERs is the detection of broad (FWHM [FORMULA] a few thousand km [FORMULA]) permitted lines in these sources, which may arise from the broad-line regions (BLR). On the analogy of the nomenclature for Seyferts, Ho (1998) has designated those sources having visible BLR as LINER 1, and others as LINER 2.

It is enlightening to see that the most part of LINER 2's in Terashima (1999) have SB notations in our Table 2, while the majority of LINERs with AGN notations in our Table 2 are listed as LINER 1's in Terashima (1999). Now we have seen a fact that these LINER 1's are basically segregated from LINER 2's in the FIR-radio diagram. This segregation confirms the early suggestion by Condon et al. (1982) of distinguishing AGN from starburst by use of the FIR-radio correlation.

The majority of sources studied in this paper are at distances of 10 Mpc or beyond, indicating that the FIR-radio diagrams shown in our figures are basically referred to the global properties of galaxies, due to the large resolution/apertures used in the NVSS and the IRAS. Nevertheless, what we have seen in the segregation is the pairing of the observed SB-supported LINERs to modeled SB-dominated passage, not in a cross-pair of the AGN- to SB-dominated passage. It would be hard to understand that the segregation could be just caused by chance.

The studies on LINERs in the IRAS 1-Jy (f(60[FORMULA] 1 Jy) sample of ultraluminous infrared galaxies (LINER ULIGs) by Veilleux et al. (1999) may shed light on the above fact. They conclude that "there is no convincing optical or infrared evidence for an AGN in LINER ULIGs", and "the main source of energy in these LINERs is a starburst rather than an AGN." We have put these LINER ULIGs in the FIR-radio diagram in Fig. 2a by symbols of crosses. Their positions are in the SB-dominated passage. Recent studies (see, Veilleux et al. 1999 and the references therein) strongly suggest that the overwhelming part of the bolometric luminosity of ULIGs stems from the inner kpc regions, indicating that the large resolution/apertures used in the NVSS and the IRAS would not make any obvious change in the FIR-radio correlation for LINER ULIGs. In fact, these studies are consistent with the early work by Kennicutt & Kent (1983), which demonstrated that in the case of EW(H[FORMULA]) [FORMULA] 10Å, suitable to LINER ULIGs, the H[FORMULA] emission observed with a slit is comparable to that obtained using large apertures.

Similar analyses of early-type galaxies (Walsh et al. 1989) suggest no obvious variance in the FIR-radio correlation with the sizes of galaxies. Radio observations of a sample of ellipticals and S0s (Fabbiano et al. 1987) have not found evidence for the extended disk emission, confirming the suggestion that the radiation is from nuclear "starburst" instead of extended disk sources (Dressel 1988). These investigations provide a sound explanation for the statistical study by Walsh et al. (1989), implying that the location of the AGN-supported LINERs (the majority of which are E/S0 sources) in the upper-left part of FIR-radio diagram (Fig. 2a) may not be significantly influenced by the size of apertures used.

On the other hand, NGC 6500, a spiral far above the border of the starburst events in Fig. 2a, has EW(H[FORMULA]) = 27Å (Ho et al. 1997), indicating little effect of the aperture sizes on the FIR-radio correlation, according to Kennicutt & Kent (1983). The same argument of EW(H[FORMULA]) [FORMULA] 10Å holds for NGC 404, one nearby galaxy, and four other LINERs selected from the Pico dos Dias Survey (PDS) (Coziol et al. 1998). Three of the PDS LINERs are classified as transition sources by Coziol et al. (1998), including NGC 3310, designated as starburst in Véron Catalog, that is classified as a transition source, SB/LINER.

The remaining sources, NGC 4736, NGC 5055, and NGC 7217, are three LINERs with EW(H[FORMULA]) [FORMULA] 10Å. For example, the EW(H[FORMULA]) of NGC 7217 obtained with a slit and a large aperture are about 3Å and 6Å  (Ho et al. 1997; Kennicutt & Kent 1983), respectively. After correcting the aperture effect over this source, the position of NGC 7217 in Fig. 2a may move a bit down- and leftward in the FIR-radio diagram, remaining in the SB-dominated passage. The same argument would be applicable to other sources with EW(H[FORMULA]) [FORMULA] 10Å.

From the above discussion, one can see that the aperture effects may not significantly change the situation of segregation for different types of LINERs shown in Fig. 2a. Therefore, as suggested by Condon et al. (1982), the FIR-radio correlation may provide a preliminary classification of LINERs according to their locations in the diagram as we described above. In other words, one may classify LINERs in terms of their FIR-to-radio ratio: [FORMULA]: one has [FORMULA] for the AGN-supported LINERs and [FORMULA] for the SB-supported ones.

In Table 3 we list part of LINERs extracted from Véron Catalog (Véron-Cetty & Véron 1996) that have detected fluxes at 1.4 GHz and 60µm, and the preliminary classification of their energetics is given in Column 6. Their distributions in the FIR-radio diagram are shown in Fig. 2b.


Table 3. LINERs from Véron Catalog

As further evidence our classification of the LINERs'types, we mention the new results of Alonso-Herrero et al. (1999). The different types of LINERs identified by these authors are consistent with our predictions for those common ones, as listed in our Table 3. For example, the claimed SB-dominated LINERs in their paper, NGC 3504, NGC 3367, NGC 4569, NGC 4826, and NGC 7743 are all located in our SB-dominated passage, and the AGN-dominated LINER NGC 2639 (Alonso-Herrero et al. 1999) is designated to be an AGN-supported LINER in Table 3. Further observations of these sources are certainly needed, especially for the LINERs located near the boundary of the starburst events shown with solid line in Fig. 2, which might be composite like NGC 6240.

3.2. LINERs with inner rings

The effects of rings or bars on Seyfert activities have been studied since the early work in the 1980's (e.g. Simkin, Su, & Schwarz 1980; Arsenault 1989). Recent studies show, however, that the frequency of barred systems is the same in Seyferts and in normal spirals (McLeod & Rieke 1995; Ho, Filippenko, & Sargent 1997; Mulchaey & Regan 1997). A latest study on the morphology of the 12µm Seyfert Sample (Hunt et al. 1999) indicates that LINERs have higher rates of inner rings than normal galaxies.

One striking feature in Fig. 2a and 2b is that almost all the LINERs having inner rings are located in the SB-dominated passage. This strong morphological tendency in the LINER sample should have important implication in starburst-AGN connection. Our preliminary analysis shows that the AGN activities of these LINERs are lower than the AGN-supported LINERs located in the upper-left part of the FIR-radio diagram. The reason for the reduced activities might be caused by the reduced fueling gas to the central black holes. We will discuss this topic in a separate paper (Lei et al. 1999), along with the morphological study of Seyferts.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: March 21, 2000