Star formation and AGN activity in the local most luminous LINERs

Star formation and AGN activity in the most luminous LINERs in the local universe

Abstract

This work presents the properties of 42 objects in the group of the most luminous, highest star formation rate LINERs at z = 0.04 - 0.11. We obtained long-slit spectroscopy of the nuclear regions for all sources, and FIR data (Herschel and IRAS) for 13 of them. We measured emission line intensities, extinction, stellar populations, stellar masses, ages, AGN luminosities, and star-formation rates. We find considerable differences from other low-redshift LINERs, in terms of extinction, and general similarity to star forming (SF) galaxies. We confirm the existence of such luminous LINERs in the local universe, after being previously detected at z  0.3 by Tommasin et al. (2012). The median stellar mass of these LINERs corresponds to 6 - 7  10 M which was found in previous work to correspond to the peak of relative growth rate of stellar populations and therefore for the highest SFRs. Other LINERs although showing similar AGN luminosities have lower SFR. We find that most of these sources have LAGN  LSF suggesting co-evolution of black hole and stellar mass. In general among local LINERs being on the main-sequence of SF galaxies is related to their AGN luminosity.

keywords:
galaxies: active; galaxies: nuclei; galaxies: star formation;
12

1 Introduction

Low Ionization Nuclear Emission line Regions (LINERs) are the most common active galactic nuclei (AGN), with numbers that exceed those of ’high ionization AGN’ (type-I and type-II Seyfert galaxies and quasars) (Heckman, 1980; Ho, 2008; Heckman & Best, 2014). At least in the local universe they make up 1/3 of all galaxies and 2/3 of AGN population (Kauffmann et al., 2003a; Yan et al., 2006; Ho, 2008). LINERs are normally classified by their narrow emission line ratios, e.g. [OIII]5007/H, [NII]6584/H, and [OI]6300/H (Baldwin et al., 1981; Kauffmann et al., 2003a; Stasińska et al., 2006; Kewley et al., 2006). In general, they have lower luminosities than Seyfert galaxies, but there is a big overlap between the groups in terms of properties like stellar mass, X-ray and radio luminosity, etc. (Ho, 2008; Netzer, 2009; Leslie et al., 2016).
Different mechanisms were proposed to explain the nature of LINERs. This includes shock excitation (e.g. Dopita et al., 1997; Nagar et al., 2005), photoionisation by young, hot, massive stars (Terlevich & Melnick, 1985), photoionisation by evolved post-asymptotic giant branch (pAGB) stars (e.g. Stasińska et al., 2008; Annibali et al., 2010; Cid-Fernandes et al., 2011; Yan & Blanton, 2012; Singh et al., 2013), and photoionisation by a central low-luminosity AGN (e.g. Ferland & Netzer, 1983; Ho, 2008; González-Martin et al., 2006). The first two proposals failed to explain the properties of large samples of LINERs. The third possibility of pAGB stars was suggested for LINERs with the weakest emission lines, located in galaxies with predominately old stars. They can be distinguished from strong-line LINERs using the equivalent widths (EW) of their emission lines, e.g., EW([OIII]5007)  1 Å (Capetti & Baldi, 2011) or EW(H)  3 Å (Cid-Fernandes et al., 2011). Several works however questioned this possibility, arguing that a population that is less luminous and more numerous than pAGB stars would be needed to produce the luminosities observed in weak LINERs (Brown et al., 2008; Rosenfield et al., 2013; Heckman & Best, 2014). However, most LINERs are powered by an AGN, especially those with stronger emission lines (e.g., EW(H)  3Å) and unresolved hard X-ray emission (e.g., González-Martin et al., 2006, 2009a, 2009b; Heckman & Best, 2014, and references therein). Like other AGNs, LINERs can be divided into type-I (broad and narrow emission lines) and type-II (only narrow emission lines). Their emission lines are characterised by lower levels of ionization than in Seyferts, and their normalized accretion rates (Eddington ratio) are 1-5 orders of magnitude smaller.
The best studied nearby LINERs (e.g. Ho, 1997, 2008; Kauffmann et al., 2003a; Leslie et al., 2016) are found in nuclei of galaxies with little or no evidence of active star formation (SF). They are usually characterised as being hosted by massive early-type galaxies (rarely spirals), and massive black holes in their centres, old stellar populations, small amounts of gas and dust, with low extinctions. Such LINERs show weak and small-scale radio jets (Ho, 2008; Heckman & Best, 2014).
Tommasin et al. (2012) studied SF in LINERs from the COSMOS field at z  0.3 using Herschel/PACS observations. They showed that: a) The SF luminosities of 34 out of 97 high luminosity LINERs are on average 2 orders of magnitude higher than SF luminosities of lower AGN luminosity, nearby LINERs. b) Even if assumed that all the observed H flux is due to SF (a wrong assumption since much of it must be due to AGN excitation) it is still impossible to recover the SF rate (SFR) indicated by the FIR observations. Given this result, we suspect that active SF in LINER host galaxies has escaped the attention of most earlier studies that focused on the innermost part of nearby galaxies. In this work we focus on the most luminous LINERs in the local (0.04  z  0.11) universe and study their SF and AGN activity, in order to understand the LINER phenomenon in relation to star-forming galaxies and to compare their properties with those of the LINERs at z  0.3. Many properties of these sources are known from SDSS spectroscopy and/or GALEX observations, e.g., emission line luminosities, locations on the BPT diagrams, SFRs based on Dn4000 estimations, etc. Unfortunately, the 3 arc-sec SDSS fibre does not allow to resolve the nuclear region and hence to separate AGN excited from SF excited emission lines. The goals of the present study are to carry out a detailed, ground based spectroscopy of the central regions of the most luminous LINERs, and to measure, together with Herschel and IRAS FIR data, their SFRs in a careful way.
The paper is organised as follows: in Section 2 we describe the sample selection. Reduction procedure for our new spectroscopic data, together with our own or archival FIR data are described in Section 3. In Section 4 we summarise all our measurements, including spectral fittings, emission line and extinction measurements, and estimations of Dn4000 and H indices, AGN luminosities, and SFRs. The main results are presented in Section 5 where we discuss the general properties of the most luminous LINERs in the local universe, co-evolution between the SF and AGN activity, and the location of our sample on the main sequence of SF galaxies.
We assumed the following cosmological parameters throughout the paper:  = 0.7,  = 0.3, and H = 70 km s Mpc.

2 Sample selection

The sources were initially selected from the SDSS/DR4 (Kauffmann et al., 2003b; Brinchmann et al., 2004) catalogue in Garching MPA-JHU based on the Sloan Digital Sky Survey (SDSS3) DR4 data (Adelman et al., 2006, and references therein). LINERs were first selected using both [NII]6584/H and [OI]6300/H criteria of Kewley et al. (2006). Taking into account the completeness of the SDSS survey, only LINERs with 0.04  z  0.11 were selected (Netzer, 2009). To eliminate LINERs ionised by pAGB stars we selected only those galaxies with H equivalent width EW(H)  2.5Å (Cid-Fernandes et al., 2011).
The next step was the selection of the most luminous LINERs within the chosen redshift interval. We measured first their AGN luminosity (LAGN) using the [OIII]5007 and [OI]6300 method of Netzer (2009) (see Section 4.4). The lines were initially corrected for reddening using the observed H/H ratio and assuming galactic extinction (see Sections 4.3 and 4.4). We selected a certain, statistically sufficient, fraction of 147 luminous LINERs with logLAGN  44.3 ergs/sec. We call these sources ’LLINERs’. Out of these sources we selected a luminosity limited sample of 47 galaxies with SF luminosity LSF  43.3 ergs/sec, where LSF is based on the Dn4000 index (see Section 4.6). Of those, we were able to obtain the optical spectra for 42 LINERs and Herschel/PACS data for 6 sources. We refer to these 42 most luminous LINERs in terms of both AGN and SF luminosity as ’MLLINERs’. All observed MLLINERs are listed in Table 1, where we provide the basic information about their properties.
Figure 1 shows the position in the LAGN vs. LSF plane of the initially classified LINERs in the selected redshift range (black dots), and the final selected sample of MLLINERs (blue squares). Using the SDSS spectroscopy we estimated the AB continuum magnitude at 6500 Å (m6500). We used these magnitudes to divide the sample into ’faint’ and ’bright’ galaxies (m6500  17.2 mag and m6500  17.2 mag, respectively). These groups are marked with F or B in Table 1. We use this classification only for observational purposes. Figs 2 and 2 (Cont.) show SDSS colour images of all MLLINERs.

Figure 1: The entire 0.04  z  0.11 SDSS/DR4 LINER sample used in this work (small black squares) and the sub-sample used for the Herschel proposal and the follow up spectroscopy (large blue squares). The dashed lines mark the lower limits on LAGN and LSF (based on Dn4000 index) used for the selection of the targets.
ID RA DEC z m6500 morph Date seeing pos. ang. texp_b texp_r Area IR data
[deg] [deg] AB [mag] [arc-sec] [deg] [sec] [sec] [arc-sec]
F01 47.499332 0.29955 0.098 18.53 S 02/11/2013 1.2 PA 3  3000.0 3  3000.0 3.6
F02 115.434586 21.18252 0.098 17.96 E 06/03/2014 1.2 314 3  3000.0 3  3000.0 3.6 1, 2
F03 131.35008 39.245438 0.109 17.63 P 08/03/2014 1.3 201 3  2000.0 3  2000.0 3.9
F04 129.59967 49.04478 0.101 17.58 P 08/03/2014 1.3 206 3  2400.0 3  2400.0 3.9
F06 144.995 34.96791 0.104 17.63 E 09/03/2014 1.4 220 3  2000.0 3  2000.0 4.2 2
F07 138.23363 46.8671 0.051 17.27 E 09/03/2014 1.4 338 3  1800.0 3  1800.0 4.2
F09 170.5683 54.6951 0.105 17.50 S 03/05/2014 1.4 149 3  2000.0 3  2000.0 4.2 2
F12 182.36954 11.030761 0.107 17.21 S 02/05/2014 1.6 410 3  2400.0 3  2400.0 6.0 2
F13 183.83566 5.533633 0.082 18.09 E 05/05/2014 1.2 120 3  3000.0 3  3000.0 3.6
F14 180.15637 4.530397 0.094 17.54 S 04/05/2014 1.2 265 3  2000.0 3  2000.0 3.6 2
F15 203.8548 45.891083 0.092 17.42 E 03/05/2014 1.4 239 3  1800.0 3  1800.0 4.2
F16 255.87796 20.849482 0.08 18.43 ? 26/07/2014 1.0 184 3  3600.0 3  3600.0 3.0 2
F17 259.5603 64.29323 0.104 17.78 E 06/03/2014 1.2 213 3  2400.0 3  2400.0 3.6 1, 2
F19 316.2105 0.358728 0.091 17.90 ? 25/07/2014 1.0 205 3  2800.0 3  2800.0 3.0 1
F20 333.30197 13.3283 0.103 18.53 P 27/07/2014 1.3 127 3  3600.0 3  3600.0 3.9
F21 342.84195 -8.956378 0.08 17.50 E 28/07/2014 1.6 241 3  3000.0 3  3000.0 4.8
F22 358.20468 14.04565 0.096 18.02 ? 29/07/2014 1.2 238 3  3200.0 3  3200.0 3.6
F23 9.282583 0.410139 0.081 17.42 ? 30/07/2014 1.4 260 3  2800.0 3  2800.0 4.2
F24 23.73075 -8.710756 0.092 18.02 P 09/10/2013 1.2 PA 3  3000.0 3  3000.0 3.6 2
B01 53.543957 1.103353 0.048 17.17 S 31/10/2013 1.5 PA 3  1600.0 3  1600.0 5.6
B02 124.66104 23.48597 0.103 16.90 P 07/03/2014 1.2 315 3  1700.0 3  1700.0 3.6 1
B03 129.57721 33.57853 0.062 16.79 P 06/03/2014 1.2 274 3  1600.0 3  1600.0 3.6 1, 2
B04 133.79796 0.219117 0.101 16.90 E 10/03/2014 1.3 255 3  1700.0 3  1700.0 3.9
B05 141.73837 8.630544 0.106 17.09 S 10/03/2014 1.3 180 3  1800.0 3  1800.0 3.9 2
B06* 160.26555 11.096189 0.053 16.50 ? 01/05/2013 0.9 PA 4  900.0 3  900.0 3.0
B07 165.55441 66.1674 0.078 17.17 P 06/03/2014 1.2 245 3  1800.0 3  1800.0 3.6 1
B08* 170.29817 -0.293878 0.098 17.11 E 03/05/2013 0.9 PA 4  1200.0 4  900.0 3.0
B09 171.66946 -1.6938 0.046 15.93 E 10/03/2014 1.3 290 3  1200.0 3  1200.0 3.9
B10 183.72675 1.916183 0.099 16.98 ? 10/03/2014 1.3 315 3  1700.0 3  1700.0 3.9
B11 187.959 58.35786 0.103 17.03 P 03/05/2014 1.4 446 3  1800.0 3  1800.0 4.2 2
B12 190.78575 1.728797 0.092 17.09 E 05/05/2014 1.2 238 3  1800.0 3  1800.0 3.6
B13* 191.979 -3.627378 0.09 16.59 S 03/05/2013 0.7 PA 2  1200.0 4  900.0 2.3 2
B14* 192.3075 15.252789 0.083 16.90 S 01/05/2013 0.7 PA 4  900.0 3  900.0 2.3
B15* 205.55083 -0.293453 0.086 17.17 E 02/05/2013 0.7 PA 3  900.0 4  900.0 2.3
B16 207.66092 53.73111 0.108 16.95 E 09/03/2014 1.4 267 3  1700.0 3  1700.0 4.2
B17 211.27605 2.771761 0.077 17.17 P 04/05/2014 1.2 180 3  1800.0 3  1800.0 3.6 2
B18 212.88733 45.28614 0.071 17.14 E 02/05/2014 1.6 109 3  1800.0 3  1800.0 6.0
B19* 230.6967 59.35285 0.076 17.09 P 01/05/2013 0.8 PA 4  1200.0 4  900.0 2.6
B20!* 231.55424 3.884864 0.086 16.79 E 02/05/2013 0.9 PA 3  1200.0 3  900.0 3.0
B21 234.29971 41.0717 0.098 16.68 E 28/07/2014 1.6 136 3  1800.0 3  1800.0 6.0
B22! 245.43016 29.725689 0.098 16.50 E 29/07/2014 1.2 264 3  1800.0 3  1800.0 3.6
B23 327.73575 -6.819708 0.059 16.68 E 26/07/2014 1.0 151 3  1800.0 3  1800.0 3.0

Column description: ID - MLLINER identification (sources observed with NOT are marked with ’*’; sources marked with ’!’ are possibly Sy2 galaxies and not LINERs as explained in Section 4.3); RA, DEC - J2000 right ascension and declination in degrees; z - redshift, from SDSS public catalogues; m6500 - AB continuum magnitude at 6500 Å; morph - visual morphological classification where E, S, and P stand for Elliptical/S0, spiral, and peculiar (see the text); Date - date of observation; seeing - average FWHM of the seeing in arc-sec; pos. ang. - slit position angle in degrees (PA means that the paralactic angle was used, otherwise the angle is orientated along the major axis); texp_b and texp_r - total exposure time in blue and red parts in seconds; Area - area covered with our ’nuclear’ extraction, in arc-sec (just for comparison, the SDSS spectra cover an area of 7.08 arc-sec); IR data - availability of Herschel (1) and IRAS (2) data.

Table 1: Summary of observations.
Figure 2: SDSS gri colour images of our selected sample of the most luminous local LINERs. The top and bottom identifications correspond to our and SDSS ones, respectively.
Figure 2 (Cont.):

3 The Data

In this section we describe the optical spectroscopic observations and data reduction that we carried out for the 42 MLLINERs. We also describe the Herschel and IRAS FIR observations used in this project. To deal with catalogues we made use of TOPCAT (Taylor, 2005), while for spectral and displaying purposes we used SIPL code (Perea J.4, priv. communication).

3.1 Optical spectroscopy

The observations were carried out during six runs (PI I. Márquez), between October 2013 and July 2014, using the Cassegrain Twin Spectrograph (TWIN) attached to the 3.5 m telescope at Calar Alto Observatory (CAHA5, Almería, Spain). Table 1 summarises the information related with observations, including the date of observation, average seeing, position angle, and exposure times. As mentioned in the previous section, we observed 42 LINERs in total. We used the T01 (red) grating during all runs, covering a spectral range of 6700 Å - 8300 Å. In the blue, we used the T08 (3500 Å - 6500 Å) grism during the first two runs (October and November 2013), and T13 (3700 Å - 7000 Å) in the following ones. The spectral sampling for T01, T08, and T13 is 0.8, 1.1, and 2.1 Å/pix, respectively. The size of the slit used is 1.2 arc-sec for seeing  1.5 arc-sec, and 1.5 arc-sec for seeing  1.5 arc-sec. The values of seeing are listed in Table 1.
Additionally, ten bright MLLINERs were observed during four nights in May 2013 (PI I. Márquez) with the Andalucía Faint Object Spectrograph and Camera (ALFOSC) of the 2.5 m telescope at the Nordic Optical Telescope (NOT6, Roque de los Muchachos Observatory, La Palma, Canary Islands, Spain). For six sources the S/N ratio was higher than for CAHA observations, and were consequently used throughout this work (marked with * in Table 1). We used #6 and #8 gratings, covering the spectral ranges 3200 Å - 5550 Å and 5825 Å - 8350 Å, in the blue and red, with a typical spectral sampling of 1.4 and 1.3 Å/pixel, respectively. We used a slit of 1.3 arc-sec in all observations. Several target exposures were taken (see Table 1) for cosmic rays and bad pixel removal. Arc lamp exposures were obtained before and after each target observation. At least two standard stars (up to four) were observed at the beginning and at the end of each night through a 10 arcsec width slit. For the final flux calibration we only considered the combination of those stars where the difference of their computed instrumental sensitivity function was lower than 10%.

Spectroscopic data reduction was carried out using IRAF7. We followed the standard steps of bias subtraction, flat-field correction, wavelength calibration, atmospheric extinction correction, and flux calibration. The sky background level was determined by taking median averages over two strips on both sides of the galaxy signal, and subtracting it from the final combined galaxy spectra. As a sanity check, we compared the reduced and calibrated spectra with the SDSS ones, scaling our data to map similar areas. Good agreement was found between the two data sets, with differences lower than 20% in both, blue and red parts of the spectra.

Morphological classification was done visually, by three independent classifiers, using the SDSS colour images shown in Figs 2 and 2 (Cont.). We separated all galaxies between early-type (E: ellipticals and lenticulars), spiral (S), and peculiar (P). The type represented in Table 1 is the one assigned by the majority of the classifiers (three or two). When the classification results in three different types, we leave the source unclassified (symbol ’?’ in the table). P class was assigned to those sources showing a clear presence of interactions, additional structures (e.g., tails, rings), and/or irregular shapes. More discussion about galaxy morphology is given in Section 5.1.

3.2 Far-infrared photometry

Herschel/PACS

We obtained FIR data for 6 objects in our sample (symbol 1 in column 13, Table 1) using the Photo detector Array Camera and Spectrometer (PACS) on board of the Herschel Space Observatory8. The data are part of a large LINER proposal (PI H. Netzer) out of which 6 targets were observed. We obtained 3  photometry with PACS blue and red bands, at 70 and 160 m, respectively. The data were processed using the standard procedure and Herschel Interactive Processing Environment (HIPE) tool (Ott et al., 2006). We extracted flux densities and their errors using again the standard HIPE tools. The fluxes and their errors are listed in Table 2.

Iras

We collected the available FIR flux measurements made by the Infrared Astronomical Satellite (IRAS9). Using the catalogue of galaxies and QSOs, Point Source Catalog (PSC), and Faint Source Catalog (FSC), we found 13 sources in total with flux densities measured or estimated as upper limits in all four IRAS bands, at 12, 25, 60 and 100 m. All these sources are listed in the last column of table 1, while the flux densities are provided in table 2. In the 60 m band, all detections have quality flag = 3 (high quality), while for the 100 m band, 10 detections have flag = 2 (moderate), and 3 sources have flag = 1 (upper limit). We only used the data with flags = 3 or = 2. For sources with flag = 1, we only used the information from the 60 m band (see Section 4.6 for more information).
Three of the IRAS observed sources (F02, F17, and B03) were also observed with Herschel/PACS. We compared the fluxes between PACS 70 m and IRAS 60 m, as well as the total SFRs measured with both surveys, and found only small differences. In the following analysis we will use the Herschel/PACS measurements for these three sources.

ID Herschel_70 Herschel_160 IRAS_60 IRAS_100
F02 0.2911  0.001 0.3909  0.0024 0.2369 (3) 1.811 (1)
F06 0.2366 (3) 0.9363 (1)
F09 0.3957 (3) 0.9608 (2)
F12 0.4728 (3) 0.9564 (2)
F14 0.3074 (3) 0.6414 (2)
F16 0.5082 (3) 0.9774 (2)
F17 0.2894  0.0036 0.2431  0.0068 0.2993 (3) 0.4687 (1)
F19 0.02  0.0011 0.0604  0.0025
F24 0.2772 (3) 0.6128 (2)
B02 0.1153  0.0037 0.1859  0.0069
B03 0.7758  0.0037 1.2186  0.007 0.784 (3) 1.356 (2)
B05 0.2821 (3) 0.8639 (2)
B07 0.1229  0.0037 0.3408  0.0069
B11 0.289 (3) 0.6007 (2)
B13 0.7087 (3) 0.8789 (2)
B17 0.5019 (3) 0.9337 (2)

Column description: ID - MLLINER identification; Herschel_70 and Herschel_160 - FIR flux and its error in the 70 m and 160 m Herschel/PACS bands, respectively, in Jy; IRAS_60 and IRAS_100 - IRAS FIR flux and the quality flag in 60 m and 100 m bans, respectively, in Jy (quality flag is given between the brackets, where 3 means high quality, 2 moderate quality, and 1 an upper limit).

Table 2: Summary of FIR observations with Herschel and IRAS.

4 Data Analysis and Measurements

4.1 Dn4000 and H measurements

Using the flux calibrated spectra, we measured the strength of 4000  break (Dn4000) and Balmer absorption-line index H. These two indices are known to be important for tracing the star formation histories (SFH) in galaxies (Kauffmann et al., 2003b). Dn4000 was measured as explained in Balogh et al. (1999), as the ratio between the average flux density in the continuum bands 4000 - 4100  and 3850 - 3950 . To obtain the H index we used the definition of Worthey and Ottaviani (1997). We first measured the average fluxes in two continuum bandpasses, blue (4041.60 - 4079.75 ), and red (4128.50 - 4161.00 ). The two average fluxes defined the continuum which we used to measure the H index, carrying out the integration within the feature in the band 4083.50 - 4122.25  and expressing it in terms of the equivalent width. Table 3 lists all these values. The main purpose of measuring Dn4000 is for using it later as a SFR indicator, while H was mainly used as an additional parameter of consistency of our measurements when comparing it with Dn4000. Previous works showed that the typical values for early-type galaxies are Dn4000  1.7 and H 1 (Kauffmann et al., 2003a).
We compared our Dn4000 and H measurements with those from the MPA-JHU DR7 database measured on SDSS spectra (Brinchmann et al., 2004). In general, for both parameters we found a good agreement between the two, with Spearman’s rank correlation coefficients p = 0.81 and 0.84, when comparing Dn4000 and H, respectively.

ID Dn4000 H ID Dn4000 H
F01 1.32  0.37 0.90 B03 3.16
F02 1.37  0.28 3.70 B04 1.47  0.31 4.67
F03 1.45  0.34 2.81 B05 1.41  0.32 3.54
F04 1.51  0.34 0.85 B06 1.30  0.22 5.59
F06 1.35  0.27 5.20 B07 1.39  0.34 1.26
F07 7.46 B08 1.43  0.24 2.07
F09 1.44  0.35 2.97 B09
F12 1.49  0.40 5.44 B10 1.28  0.23 4.10
F13 5.59 B11 1.26  0.15 3.54
F14 1.37  0.30 4.16 B12 1.33  0.28 0.83
F15 1.33  0.27 4.93 B13 1.01  0.12 4.52
F16 1.16  0.25 0.66 B14 1.52  0.26 1.28
F17 1.26  0.29 7.80 B15 1.27  0.25 7.04
F19 1.39  0.33 0.24 B16 1.36  0.30 6.23
F20 1.16  0.20 2.25 B17 1.21  0.27 6.90
F21 1.37  0.30 5.98 B18 1.34  0.26 6.00
F22 1.17  0.18 4.94 B19 1.32  0.27 7.21
F23 1.30  0.27 6.48 B20! 1.42  0.27 3.20
F24 1.32  0.56 8.26 B21 1.45  0.27 0.24
B01 1.23  0.29 5.49 B22! 1.22  0.23 5.79
B02 1.16  0.22 6.35 B23 1.46  0.27 5.96

! possibly Sy2 galaxies (see Section 4.3)

Table 3: Dn4000 and H measurements.

Figure 3 shows the relation between the Dn4000 and H indices obtained by Kauffmann et al. (2003b) for the SDSS DR4 sample (see their figure 6). They used a library of 32,000 different SFH, where for each SFH they have a corresponding Dn4000 and H indices, as well as the fraction of the total stellar mass of the galaxy formed in the bursts over the past 2 Gyr (F). In their figure the bins are coded according to the fraction of model SFHs with F in a given range (see the caption of their Fig. 3). We used this figure and overplotted our Dn4000 and H measurements (coloured filled circles). In general our measurements are consistent with the models by Kauffmann et al. (2003b). More details about star formation histories are given in Section 5.1.

Figure 3: Figure 6 of Kauffmann et al. (2003b), showing the relation between the Dn4000 and H indices for their sample of SDSS sources, with our sample over-plotted (colour filled circles). Open and solid triangles show the indices with high confidence that the galaxy has experienced a burst over the past 2 Gyr (our green and yellow circles, respectively). Open (solid) triangles indicate regions where 95% of the model galaxies have F 0.05 and the burst occurred more than (less than) 0.1 Gyr ago. Solid squares indicate regions where 95% of the model galaxies have F 0 (our orange circle). Regions marked with crosses contain a mix of bursty and continuous star formation models (our red circles). Violet circles lie outside the range covered by models.

4.2 STARLIGHT spectral fittings of the nuclear regions

We extracted what we call the nuclear spectra by selecting a central region equal to 2.5 times the FWHM of the seeing. In the case of CAHA the extraction size is 5 -8 pixels (depending on the used slit), while in the case of NOT data the central 10 pixels were extracted. The total area covered by the nuclear extraction is given in the last column of Table 1 for all MLLINERs.
Modelling of the nuclear stellar spectra of our sources was performed with the STARLIGHT10 V.04 synthesis code (Cid-Fernandes et al., 2005, 2009). All spectra were previously corrected for galactic extinction, K-corrected, and moved to rest-frame. To correct for the galactic extinction we used the pystarlight11 library within the astrophysics Python package12 and Schlegel et al. (1998) maps of dust IR emission. Our fittings are based on the templates from Bruzual & Charlot (2003), with solar metallicity and 25 different stellar ages, from 0.001  10 to 18  10. Considering that we are dealing with nuclear spectra of large galaxies, this approximation should be fine for our sources (Ho, 2003, 2008). We masked in all spectra the emission line regions, areas with atmospheric absorptions, and regions with bad pixels. To measure the signal-to-noise (S/N) we checked visually all spectra to select the continuum region free of bad pixels, using always the blue range and a width of at least 80Å. In most cases we selected the region around  4600Å  or  5600Å. S/N measurements are listed in Table 4.
In this work we used the Cardelli et al. (1989) extinction law. This law was widely used in different surveys for fitting the host-dominated sources (Stasińska et al., 2006; Cid-Fernandes et al., 2011; González Delgado et al., 2015). We also tested the Calzetti et al. (2007) law, and made a comparison between the two. We found differences to be lower than 20% therefore we only show results that were obtained with the Cardelli et al (1989) extinction law.

The basic information obtained from the best fit stellar population models is summarised in Table 4 for all MLLINERs. The adev parameter gives the goodness of the fit, and presents the mean deviation over the all fitted pixels (in percentage); adev  6 and  10 stand for ’very good’ and ’good’ fits, respectively (Cid-Fernandes et al., 2005, 2009). We obtained very good fit in  80% of the cases. The measured S/N ratio and extinction A are given in columns 3 and 4, respectively. The best fit parameters (M_cor_tot and M_ini_tot) were used to measure two types of stellar masses, following Cid-Fernandes et al. (2005, 2009):
the present mass in stars,

M = M_cor_tot  10 4d (3.826  10),

and the initial mass, that has been processed into stars throughout the galaxy life:

M = M_ini_tot  10 4d (3.826  10).

The results regarding the best stellar population mixture are summarised in columns 7 - 9. They are represented through the light-fraction population vector (corresponds to the same wavelength selected for measuring S/N, see above) for three stellar ages: young (with age [yr]  10), intermediate (10 age [yr]  10), and old (age [yr]  10). We discuss stellar populations in more detail in Section 5.1. Finally, we calculated the light-weighted mean ages of our MLLINERs, using as a reference Cid-Fernandes et al. (2013):

logt = logt,

where is a fraction of light at stellar age t in our best-fit model and metallicity Z (Z in our case). The example with the best-fit models (red lines) and original spectra (blue lines) are shown in Appendix A (Figs. 13).

ID adev S/N A M M stpop1 stpop2 stpop3 logt ID adev S/N A M M stpop1 stpop2 stpop3 logt
F01 9.62 15.16 0.138 0.76 1.46 4.83 22.08 71.42 9.03 B03 2.64 44.51 0.877 2.89 5.79 15.81 14.67 67.75 9.19
F02 4.70 21.69 1.667 2.50 4.70 15.17 18.92 66.47 8.97 B04 3.54 44.18 0.584 1.93 3.41 0.0 0.0 93.51 8.61
F03 4.59 25.82 0.584 1.67 3.12 6.38 0.0 95.8 9.53 B05 4.39 30.96 0.753 3.77 7.20 1.72 3.4 88.68 8.77
F04 4.27 26.74 0.827 2.17 4.07 6.89 0.0 97.81 9.82 B06 2.68 45.25 1.209 3.35 6.65 0.0 55.28 42.89 8.95
F06 3.74 36.17 1.306 1.56 2.64 5.64 10.51 79.16 8.51 B07 4.26 37.31 2.209 6.81 13.63 8.55 20.57 70.08 9.46
F07 3.37 27.42 1.162 0.61 1.21 41.52 42.3 18.26 8.17 B08 3.43 31.36 0.388 4.64 9.29 0.0 53.01 49.76 9.45
F09 8.49 18.36 1.423 1.93 3.44 0.0 10.96 90.84 9.36 B09 2.60 44.41 0.234 0.96 1.77 9.18 0.0 87.64 8.97
F12 5.21 31.79 0.459 0.86 1.56 0.0 43.09 54.45 9.06 B10 3.23 47.73 0.787 1.19 2.02 5.21 12.4 78.21 8.59
F13 12.60 23.90 1.515 0.58 0.99 0.0 4.24 90.83 8.71 B11 3.43 43.65 0.517 1.68 2.95 0.0 14.94 75.46 8.26
F14 7.53 20.31 0.8 1.10 2.04 0.0 89.94 8.7 8.94 B12 5.66 27.73 0.706 0.57 0.97 0.0 24.78 70.88 8.7
F15 5.09 28.67 0.991 1.05 1.87 0.0 96.77 3.39 8.99 B13 2.17 50.33 0.673 1.44 2.47 7.65 65.0 26.36 8.18
F16 7.51 17.07 1.239 0.88 1.70 16.44 12.77 69.52 8.69 B14 3.39 41.95 0.741 4.75 9.42 0.0 25.8 74.38 9.5
F17 3.55 43.21 1.369 1.02 1.71 7.1 32.41 58.34 8.58 B15 2.89 43.62 0.701 1.03 1.85 0.0 92.53 9.72 9.0
F19 7.39 18.17 0.542 1.54 2.98 0.0 17.79 83.21 9.68 B16 3.49 39.68 0.327 1.33 2.31 0.0 0.0 98.71 9.07
F20 6.51 22.11 1.154 0.90 1.69 6.37 47.45 46.19 8.68 B17 3.95 45.53 0.997 1.47 2.83 0.0 57.78 42.84 9.07
F21 7.49 22.74 0.738 1.52 2.84 0.0 49.74 50.4 9.27 B18 4.38 32.81 0.916 1.44 2.78 0.0 78.52 18.81 8.88
F22 4.30 30.01 0.698 1.45 2.81 0.0 49.72 48.15 8.75 B19 4.48 33.10 1.914 5.61 11.06 0.0 77.21 21.86 8.76
F23 5.25 34.38 0.702 1.12 2.13 0.0 86.75 9.27 8.58 B20! 3.20 31.32 0.312 5.41 10.82 0.0 51.55 53.01 9.9
F24 11.55 10.97 1.68 1.78 3.52 12.06 57.86 31.98 8.98 B21 4.41 51.15 0.515 8.32 16.65 0.0 38.45 56.92 9.28
B01 5.67 23.82 0.681 0.69 1.36 2.67 74.6 23.11 8.72 B22! 3.71 42.86 0.462 4.85 9.52 0.0 57.46 36.84 8.69
B02 2.87 41.10 1.105 3.89 7.65 9.67 55.18 32.08 8.45 B23 4.26 35.50 1.73 3.31 0.0 63.79 34.14 9.11

Column description: ID - MLLINER identification (’!’ - possibly Sy2 galaxies, see Section 4.3); adev - goodness of the fit (see the text); S/N - measured signal-to-noise ratio (see Section 4.2); A - extinction in V band; M and M - current and initial mass in stars, respectively, in 10 [M]; stpop1 - fraction of young stars with age [yr]  10 in %; stpop2 - fraction of intermediate stars with 10 age [yr]  10 in %; stpop3 - fraction of old stars with age [yr]  10 in %; logt - mean age.

Table 4: The best stellar population mixture found by STARLIGHT.

4.3 Emission line measurements and classification

We obtained the nuclear emission line spectra by subtracting from the original ones the best-fit stellar models. Fig 13 - LABEL:fig_spectra_5 show the final emission spectra (black solid lines) of all MLLINERs. We used these spectra to measure the properties of the emission lines. Using splot IRAF task, we measured the flux of the strong emission lines by fitting a single gaussian function. Table 5 summarises the resulting fluxes for [OII]3727, H, [OIII]4959, [OIII]5007, [OI]6300, [NII]6548, [NII]6584, [SII]6718, and [SII]6731 lines relative to the H line. The errors were measured taking into account the rms of the continuum. We also measured the equivalent width (EW) of H line by fitting again the line with a single Gaussian function and using the original spectra.
All emission lines were corrected for extinction using the ratio of HI Balmer lines, and using H/H = 3.1 as the theoretical value for AGN (Osterbrock & Ferland, 2005). Table 6 summarises the corrected flux ratios, again relative to the H line. We also summarise the measured values of extinction in the V band (A). We compared these values with those obtained from the STARLIGHT best-fit models (see Section 4.2 and Table 4), finding in general important discrepancies between the two measurements, where the emission-line technique gives in general higher values of A, as has been seen previously (Calzetti et al., 1994).

ID [OII]3272 H [OIII]4959 [OIII]5007 [OI]6300 [NII]6548 [NII]6584 [SII]6716 [SII]6731 F 10 EW
F01 0.61  0.19 0.24  0.03 0.17  0.04 0.42  0.04 0.35   0.06 1.29  0.09 0.25  0.08 0.17  0.08 14.41  0.77 17.1
F02 0.11  0.03 0.14  0.03 0.27   0.06 0.93  0.08 0.24  0.05 0.2   0.05 14.31  0.88 14.46
F03 0.23  0.05 0.14  0.06 0.41  0.07 0.38   0.09 0.84  0.11 0.54  0.09 0.37  0.08 14.27  1.2 11.44
F04 0.17  0.04 0.07  0.02 0.31  0.05 0.41   0.09 1.12  0.12 0.53  0.11 0.39  0.1 10.44  0.86 6.79
F06 0.17  0.03 0.04  0.03 0.29  0.03 0.3   0.05 0.93  0.07 0.24  0.04 0.19  0.04 20.82  1.0 10.37
F07 0.12  0.06 0.24  0.08 0.66  0.13 0.42   0.19 1.66  0.34 0.33  0.08 0.23  0.06 13.76  2.42 7.99
F09 0.44  0.2 0.19  0.1 0.2   0.08 0.38   0.23 1.67  0.41 0.45  0.24 0.39  0.24 5.3   1.12 1.64
F12 0.39  0.24 0.12  0.07 0.13  0.09 0.23   0.49 1.03  0.69 0.27  0.14 0.23  0.13 10.83  5.17 8.96
F13 0.3   0.1 0.14  0.04 0.1   0.05 0.26  0.07 0.39   0.08 1.12  0.11 0.39  0.1 0.33  0.1 8.74   0.65 8.9
F14 0.89  0.14 0.27  0.06 0.1   "" 0.35  0.07 0.68   0.12 1.71  0.19 0.57  0.11 0.47  0.11 16.84  1.65 10.1
F15 0.49  0.06 0.13  0.03 0.08  0.05 0.27  0.05 0.52   0.09 1.1   0.12 0.4   0.07 0.27  0.07 18.4   1.53 6.83
F16 0.27  0.02 0.14  0.01 0.06  0.01 0.2   0.01 0.34   0.02 0.86  0.03 0.18  0.02 0.16  0.02 34.68  0.7 44.99
F17 0.17  0.03 0.07  0.02 0.36  0.04 0.43   0.06 1.05  0.09 0.22  0.04 0.18  0.04 22.58  1.33 16.12
F19 0.88  0.07 0.21  0.02 0.22  0.03 0.54  0.04 0.26   0.05 0.65  0.06 0.46  0.06 0.39  0.06 21.56  1.13 17.21
F20 0.4   0.03 0.2   0.02 0.15  0.02 0.34   0.03 0.83  0.04 0.27  0.02 0.19  0.02 26.29  0.81 36.03
F21 0.61  0.08 0.13  0.05 0.41  0.05 0.41  0.05 0.56   0.08 1.86  0.15 0.48  0.09 0.39  0.09 24.27  1.75 13.47
F22 0.42  0.06 0.26  0.02 0.08  0.03 0.19  0.03 0.25   0.04 1.13  0.05 0.26  0.05 0.2   0.05 23.82  0.81 13.58
F23 0.92  0.21 0.32  0.34 0.2   0.1 0.62  0.13 0.47   0.13 1.31  0.2 6.06   0.72 1.43
F24 0.48  0.21 0.37   0.17 0.7   0.2 0.66  0.25 0.43  0.24 3.59   0.58 6.11
B01 0.65  0.07 0.19  0.02 0.07  0.02 0.22  0.02 0.43   0.04 1.07  0.06 0.42  0.03 0.35  0.02 61.67  2.46 24.33
B02 0.22  0.04 0.07  0.03 0.3   0.07 0.38   0.06 0.97  0.08 0.29  0.08 0.27  0.08 13.63  0.82 4.87
B03 0.11  0.01 0.05  0.01 0.15  0.01 0.11  0.01 0.3   0.03 0.93  0.04 0.23  0.02 0.22  0.02 114.2  2.94 40.9
B04 0.22  0.17 0.21  0.09 0.55  0.14 0.88   0.26 1.75  0.4 11.47  2.28 1.48
B05 0.17  0.04 0.05  0.03 0.14  0.05 0.43   0.07 1.27  0.1 0.19  0.05 0.18  0.05 18.95  1.23 7.01
B06 0.95  0.08 0.25  0.05 0.07  0.05 0.33  0.04 0.27  0.05 0.21   0.05 1.23  0.08 0.54  0.07 0.41  0.07 28.11  1.49 4.84
B07 0.1   0.03 0.14  0.04 0.31   0.07 0.93  0.09 0.43  0.1 0.33  0.09 12.43  0.78 6.94
B08 1.27  0.1 0.24  0.06 0.13  0.04 0.4   0.04 0.32  0.2 0.5   0.08 1.56  0.13 0.75  0.11 0.61  0.1 21.48  1.53 9.07
B09 0.37  0.05 0.47  0.06 1.57  0.13 0.36  0.05 0.32   0.08 0.7   0.09 0.74  0.08 0.59  0.07 55.34  4.09 8.37
B10 0.19  0.08 0.16  0.06 0.38   0.08 1.08  0.11 0.32  0.09 0.24  0.09 19.84  1.42 7.18
B11 0.63  0.04 0.22  0.02 0.05  0.02 0.2   0.02 0.44   0.05 0.97  0.06 0.46  0.03 0.35  0.03 43.09  1.81 15.51
B12 0.64  0.11 0.24  0.04 0.25  0.04 1.14  0.08 0.4   0.06 0.92  0.07 0.51  0.11 0.39  0.11 17.07  0.93 10.37
B13 0.4   0.02 0.19  0.01 0.05  0.0 0.18  0.01 0.29   0.02 0.68  0.02 0.19  0.02 0.17  0.02 223.1  3.95 47.02
B14 0.16  0.04 0.26  0.06 0.05  0.01 0.28  0.04 1.07   0.17 0.59  0.14 44.97  5.35 15.14
B15 0.81  0.07 0.21  0.04 0.05  0.02 0.29  0.04 0.24  0.08 0.53   0.08 1.63  0.13 0.39  0.07 0.32  0.07 29.15  1.99 10.58
B16 0.19  0.05 0.22  0.06 0.53  0.12 0.52   0.11 1.4   0.16 0.61  0.13 0.5   0.13 12.39  1.16 3.93
B17 0.21  0.07 0.1   0.02 0.1   0.05 0.22  0.03 0.31   0.07 1.05  0.09 0.41  0.12 0.21  0.11 23.88  1.49 8.93
B18 0.36  0.08 0.25  0.07 0.19  0.07 0.5   0.1 1.19  0.14 0.12  0.05 0.18  0.05 18.32  1.62 6.21
B19 0.26  0.05 0.18  0.04 0.04  0.01 0.33  0.05 0.19  0.07 0.46   0.06 1.36  0.09 0.39  0.08 0.32  0.08 15.75  0.86 4.32
B20 0.99  0.06 0.25  0.03 0.57  0.04 1.64  0.09 0.35  0.06 0.53   0.05 1.54  0.08 0.48  0.06 0.42  0.06 32.83  1.49 7.41
B21 0.6   0.18 0.14  0.07 0.42  0.13 1.66  0.41 0.29  0.16 0.15  0.15 16.94  3.59 3.57
B22 0.81  0.2 0.19  0.11 0.68  0.16 2.04  0.41 0.52   0.22 2.63  0.55 0.41  0.12 0.44  0.12 17.63  3.43 3.05
B23 0.32  0.03 0.13  0.01 0.08  0.02 0.32  0.02 0.36   0.04 0.84  0.04 0.38  0.05 0.31  0.05 63.83  2.19 16.05

Column description: ID - MLLINER identification (’!’ - possibly Sy2 galaxies, see Sec. 4.3); [OII]3272, H, [OIII]4959, [OIII]5007, [OI]6300, [NII]6548, [NII]6584, [SII]6716, and [SII]6731 - ratio between the fluxes of the indicated emission lines and H line; F 10 - flux of the H line in [erg/cm/sec]; EW - H equivalent width measured in the spectrum before the subtraction of the best model for the underlying stellar population.

Table 5: Properties of strong emission lines.
ID [OII]3272 [OIII]4959 [OIII]5007 [OI]6300 [NII]6548 [NII]6584 [SII]6716 [SII]6731 F 10 L 10 A
F01 1.09  0.34 0.22  0.05 0.54  0.06 0.35  0.06 1.29  0.09 0.25  0.08 0.17  0.08 28.68   6.67 7.03   1.63 0.872
F02 0.36  0.11 0.27  0.08 0.93  0.19 0.24  0.07 0.2   0.06 177.85   49.69 43.74   12.22 3.191
F03 0.18  0.08 0.55  0.1 0.38  0.09 0.84  0.12 0.54  0.09 0.37  0.08 32.09   9.38 9.74   2.85 1.026
F04 0.13  0.05 0.55  0.12 0.41  0.1 1.12  0.17 0.53  0.12 0.39  0.11 48.73   14.38 12.69   3.75 1.951
F06 0.07  0.06 0.53  0.08 0.3   0.05 0.93  0.09 0.24  0.05 0.19  0.04 99.64   22.31 27.69   6.2 1.983
F07 0.57  0.3 1.54  0.73 0.42  0.23 1.66  0.63 0.33  0.13 0.23  0.1 134.56   63.88 8.36   3.97 2.888
F09 1.22  0.7 0.32  0.14 0.38  0.24 1.67  0.52 0.45  0.26 0.39  0.25 17.69   8.5 4.96   2.38 1.528
F12 2.64  3.66 0.29  0.34 0.23  0.51 1.03  0.98 0.27  0.23 0.23  0.2 103.78   87.17 30.5   25.61 2.862
F13 1.5   0.61 0.2   0.11 0.53  0.18 0.39  0.1 1.12  0.19 0.39  0.12 0.33  0.11 58.59   16.99 9.72   2.82 2.41
F14 1.23  0.21 0.4   0.09 0.68  0.12 1.71  0.2 0.57  0.11 0.47  0.11 24.86   7.8 5.51   1.73 0.493
F15 3.16  0.95 0.18  0.11 0.61  0.17 0.52  0.12 1.1   0.2 0.4   0.09 0.27  0.08 167.08   51.2 35.22   10.79 2.794
F16 1.47  0.16 0.12  0.03 0.43  0.03 0.34  0.02 0.86  0.04 0.18  0.02 0.16  0.02 255.32   36.87 40.48   5.84 2.528
F17 0.11  0.03 0.64  0.09 0.43  0.07 1.05  0.11 0.22  0.05 0.18  0.05 101.99   25.28 27.86   6.91 1.91
F19 1.99  0.2 0.31  0.04 0.78  0.07 0.26  0.06 0.65  0.07 0.46  0.06 0.39  0.06 57.04   13.16 11.85   2.73 1.232
F20 1.03  0.11 0.22  0.03 0.34  0.03 0.83  0.05 0.27  0.03 0.19  0.02 79.76   14.12 21.46   3.8 1.406
F21 3.56  1.23 0.9   0.24 0.9   0.24 0.56  0.13 1.86  0.36 0.48  0.12 0.39  0.11 195.11   57.69 30.95   9.15 2.64
F22 0.65  0.09 0.09  0.04 0.22  0.04 0.25  0.04 1.13  0.05 0.26  0.05 0.2   0.05 39.57   7.31 9.14   1.69 0.643
F23 0.91  0.2 0.19  0.1 0.62  0.13 0.47  0.13 1.31  0.2 6.01   2.06 0.96   0.33 -0.011
B01 1.95  0.26 0.12  0.03 0.36  0.04 0.43  0.05 1.07  0.07 0.42  0.03 0.35  0.03 227.37   45.83 12.22   2.46 1.652
B02 0.09  0.05 0.43  0.1 0.38  0.07 0.97  0.1 0.29  0.08 0.27  0.08 35.35   8.75 9.55   2.36 1.207
B03 0.11  0.03 0.37  0.04 0.13  0.01 0.3   0.03 0.93  0.06 0.23  0.03 0.22  0.03 1359.55  223.14 126.89  20.83 3.137
B04 0.28  0.14 0.75  0.26 0.88  0.3 1.75  0.49 27.04   12.43 7.03   3.23 1.086
B05 0.09  0.05 0.25  0.09 0.43  0.08 1.27  0.15 0.19  0.05 0.18  0.05 84.22   22.05 24.24   6.35 1.889
B06 1.57  0.16 0.09  0.06 0.41  0.05 0.28  0.05 0.21  0.05 1.23  0.09 0.54  0.07 0.41  0.07 50.85   11.73 3.39   0.78 0.751
B07 0.4   0.15 0.31  0.09 0.93  0.2 0.43  0.13 0.33  0.11 190.8   54.47 28.31   8.08 3.459
B08 2.24  0.24 0.17  0.05 0.52  0.06 0.34  0.21 0.5   0.08 1.56  0.15 0.75  0.11 0.61  0.11 42.03   11.27 10.3   2.76 0.85
B09 0.42  0.05 1.39  0.12 0.35  0.05 0.32  0.08 0.7   0.09 0.74  0.08 0.59  0.07 39.99   10.88 2.03   0.55 -0.412
B10 0.26  0.11 0.38  0.09 1.08  0.16 0.32  0.1 0.24  0.09 72.72   20.36 17.98   5.03 1.645
B11 1.42  0.12 0.07  0.03 0.28  0.03 0.44  0.05 0.97  0.06 0.46  0.03 0.35  0.03 112.47   23.18 30.65   6.32 1.215
B13 1.21  0.08 0.08  0.01 0.28  0.01 0.29  0.02 0.68  0.02 0.19  0.02 0.17  0.02 819.2   109.52 167.61  22.41 1.647
B14 0.25  0.06 0.06  0.02 0.34  0.05 1.07  0.18 0.59  0.14 74.49   25.78 12.79   4.43 0.639
B15 1.93  0.26 0.07  0.03 0.42  0.07 0.25  0.09 0.53  0.08 1.63  0.16 0.39  0.07 0.32  0.07 81.78   21.61 15.11   3.99 1.306
B16 0.35  0.1 0.82  0.21 0.52  0.11 1.4   0.2 0.61  0.14 0.5   0.14 40.88   12.76 12.26   3.82 1.512
B17 2.07  0.9 0.28  0.16 0.59  0.16 0.31  0.08 1.05  0.19 0.41  0.13 0.21  0.12 354.13   96.75 51.23   14.0 3.415
B18 0.59  0.13 0.23  0.09 0.5   0.1 1.19  0.15 0.12  0.05 0.18  0.05 32.35   9.65 3.99   1.19 0.72
B19 0.82  0.19 0.06  0.02 0.55  0.09 0.21  0.07 0.46  0.07 1.36  0.14 0.39  0.09 0.32  0.08 62.04   14.86 8.85   2.12 1.736
B20! 1.68  0.13 0.72  0.06 2.07  0.13 0.37  0.06 0.53  0.05 1.54  0.09 0.48  0.06 0.42  0.06 61.76   13.21 11.34   2.43 0.8
B21 3.07  1.93 0.86  0.44 1.66  0.65 0.29  0.18 0.15  0.16 117.96   59.8 28.62   14.51 2.458
B22! 2.44  1.12 1.11  0.41 3.32  1.16 0.52  0.24 2.63  0.77 0.41  0.15 0.44  0.15 65.01   30.24 15.84   7.37 1.653
B23 1.95  0.28 0.18  0.04 0.72  0.08 0.36  0.04 0.84  0.07 0.38  0.05 0.31  0.05 552.23   104.81 45.77   8.69 2.733

Column description: ID - MLLINER identification (’!’ - possibly Sy2 galaxies, see Section 4.3); [OII]3272, [OIII]4959, [OIII]5007, [OI]6300, [NII]6548, [NII]6584, [SII]6716, and [SII]6731 - ratio between the extinction corrected fluxes of the indicated emission lines and H line; F 10 - flux of the extinction corrected H line in [erg/cm/sec]; L 10 - luminosity of the extinction corrected H line in [erg/sec]; A - interstellar extinction parameter in the V band in [mag].

Table 6: Properties of extinction corrected strong emission lines.
Figure 4: (Top) The BPT-NII (left), BPT-OI (centre), and BPT-SII (right) diagrams. In the BPT-NII plot black dashed line (Kauffmann et al., 2003b) and blue solid line (Kewley et al., 2001) separate HII regions and AGNs, while green dotted line (Cid-Fernandes et al., 2010) separates Seyfert (above) and LINERs (below). The BPT-OI and BPT-SII diagrams use Kewley et al. (2006) limits to distinguish between different sources. In all plots red and blue filled circles show possible outliers (see the text) classified as Seyfert and transit, respectively. The median error bars are given in all plots in the bottom left corner. (Bottom) Same as above, but using SDSS MPA-JHU DR7 data. In the BPT-NII and BPT-SII diagrams, red and blue filled circles show the position of sources being in the Seyfert and transition areas, respectively, in our plots. In the BPT-OI diagram there are fewer sources due to the availability of [OI]6300 line. We marked the position of all sources for we have data with dark blue filled circles.

Fig. 4 (top plots) shows three standard BPT diagrams based on [NII]6584, [OI]6300, and [SII]6716+6731 emission line ratios, used to separate between the HII regions and AGN, and between Seyfert 2 galaxies and LINERs (Baldwin et al., 1981). The lines correspond to Kewley et al. (2001), Kauffmann et al. (2003b), Kewley et al. (2006), and Cid-Fernandes et al. (2010) (see the caption of Fig. 4). Following the BPT-NII diagram, 4 MLLINERs (B22, B20, B09, and F19) enter in the region of Seyfert galaxies (although they are located close to the limiting line with LINERs), while another three sources (B13, B14, and F20) stay inside the transition region. As for [OI]6300 we could not detect this line in most of our CAHA observations, and we only have 7 sources plotted in the BPT-OI diagram (see Table 5). All these sources enter in the region typical of LINERs. In the BPT-SII diagram again four sources are located inside the area typical of Seyfert (B20, B21, B22, and F07), while 13 lie in the transition region (in particular B03, B05, B10, B13, B18, F01, F02, F06, F12, F16, F17, F20, and F22). We considered as possible outliers those sources that at least in two of the BPT diagrams lie outside of the standard LINER region. We found two possible Seyfert galaxies (B20 and B22, marked in red), and two possible transition galaxies (F20 and B13, marked in blue).
The discrepant classification of some of the sources is not surprising given that the original sample selection was based on the SDSS MPA-JHU DR4 data. Since then both SDSS data calibration and the analysis by the MPA-JHU have improved. We compared the positions of our MLLINERs on the BPT diagrams using the new version of MPA-JHU catalogues based on the SDSS DR7 data. These plots are presented in Fig. 4 (bottom diagrams). B20 and B22 enter in the Seyfert region in this case as well. Therefore, we will consider these two galaxies as outliers, and although we provide their measurements in all tables, we exclude them from all diagrams showed in Section 5. In all tables these two galaxies are marked with ’!’. F20 and B13 stay inside the LINER region in the SDSS DR7, so we do not consider them as outliers.

To test in more detail the nuclear classification of our MLLINERs, we also used the WHAN diagram by Cid-Fernandes et al. (2011). This diagram shows the relation between EW(H) and the [NII]6584/H ratio, and separates all galaxies in passive (lineless), retired, pure star-forming, and strong and weak AGN. The purpose is to distinguish ’true’ from ’fake’ AGN, and to separate between the two classes that overlap in the LINER region of the traditional diagnostic diagrams: galaxies hosting weak AGN, and retired galaxies that have stopped forming stars and are ionised by hot low-mass evolved (pAGB) stars. Fig. 5 represents the WHAN diagram for all our MLLINERs. Thirty three sources occupy the region of strong AGN having EW(H)  6Å and [NII]6584/H -0.4, while six (B02, B06, B16, B19, B21, and B22) are located in the area of weak AGN with 3Å  EW(H)  6Å and [NII]6584/H -0.4.
Three MLLINERs (B04, F09, and F23) show EW(H) between 1.4 and 3A and are therefore located in the area of retired galaxies. The lower limit of EW(H) = 3Å was determined by Cid-Fernandes et al. (2011) using the 3 arc-sec SDSS fibre spectra. In our case, however, we used the nuclear spectra covering a smaller area for all MLLINERs. We compared our EW(H) measurements with those of SDSS MPA-JHU DR713 finding a linear correlation, but higher MPA-JHU values in all cases due to aperture differences. Since in the MPA-JHU DR7 catalogue all our sources have EW(H)  3Å (as in the initially used DR4 version), we will continue to consider the entire selected sample as AGN, including the three sources that enter in the area of retired galaxies with the values from our nuclear spectra.

Figure 5: The revised WHAN classification diagram showing the relation between EW(H) and [NII]/H. The limits are those suggested by Cid-Fernandes et al. (2011). They are used to separate between SF galaxies, strong AGN (sAGN), weak AGN (wAGN), retired, and passive galaxies, as marked on the diagram.

4.4 AGN luminosity

Our measurements of LAGN are based on the reddening-corrected luminosity of H and [OIII]5007. We used eq. 4 from Tommasin et al. (2012) which is based on Netzer (2009):

logLAGN = logL(H) + 3.75 + max[0,0.31  (log([OIII]5007/H)-0.6)].

Table 8 summarises the obtained values for all MLLINERs. Netzer (2009) showed that [OIII]5007 and [OI]6300 lines provide more accurate measurements of the LAGN, measured as:

logLAGN = 3.8 + 0.25logL([OIII]5007) + 0.75logL([OI]6300).

However, since [OI]6300 is missing in most of our spectra (see table 6), we are able to measure the LAGN based on this line only in the case of seven MLLINERs, and therefore for consistency we won’t use these measurements in our analyses.

4.5 AGN and SF contributions to the emission lines

Both H and [OII]3727 lines can be used to estimate SFRs in non-active galaxies (Kennicutt, 1992; Kewley et al., 2004; Mouhcine et al., 2005; Moustakas et al., 2006). However, all our sources are classified as LINERs and hence much of the flux in these two lines can be due to ionization and excitation by the central non-stellar source. To assess the various contributions to L(H), we made an estimate of the expected H luminosity (using Netzer (2013) expression) based on the SFRs, obtained from STARLIGHT by using stellar absorption spectra and young stellar populations (age  10 yr). We compared these values with the measured H luminosities (see Tab. 6). Tab. 7 gives all measurements and estimated AGN contributions. For those MLLINERs without young stellar populations detected (see table 4) we assume that all H emission comes from the AGN. In almost all MLLINERs, most of the nuclear Ha is due to the AGN (all sources except 4 have AGN contribution of  60%). Therefore we do not consider the estimators based on H and [OII]3727 lines as reliable tracers of SF in our case.

ID L(H)_test AGN ID L(H)_test AGN
 10 [erg/s] [%]  10 [erg/s] [%]
F01 0.23 96.71 B03 6.49 94.87
F02 3.33 92.37 B04 0.0 100.0
F03 1.65 82.97 B05 1.69 92.99
F04 1.65 86.97 B06 0.0 100.0
F06 5.99 78.35 B07 9.83 65.25
F07 3.76 55.01 B08 0.0 100.0
F09 0.0 100.0 B09 1.08 46.37
F12 0.0 100.0 B10 2.48 86.18
F13 0.0 100.0 B11 0.0 100.0
F14 5.19 5.702 B13 4.55 97.28
F15 0.0 100.0 B14 0.0 100.0
F16 1.25 96.89 B15 0.0 100.0
F17 6.35 77.18 B16 0.0 100.0
F19 0.0 100.0 B17 0.0 100.0
F20 0.95 95.57 B18 0.0 100.0
F21 0.0 100.0 B19 0.0 100.0
F22 0.0 100.0 B20! 0.0 100.0
F23 0.0 100.0 B21 0.0 100.0
B01 0.62 94.93 B22! 0.0 100.0
B02 11.1 -16.9 B23 0.0 100.0

Column description: ID - MLLINER identification; L(H)_test - H luminosity obtained from the STARLIGHT SFRs correspondent only to young stellar populations; AGN - approximation of AGN contribution to H luminosity in %.

Table 7: AGN contribution measured through L(H) and STARLIGHT SFRs (obtained from young stellar populations).

4.6 Star formation rates

We measured the SFRs using different methods and data, both optical and FIR. In the following, we provide a full description for each measurement, and list the results in Table 8. To convert SFR to LSF, we assume a slightly rounded value of LSF  SFR  10 L based on the Kroupa initial mass function (IMF). When scaling the nuclear measurements of SFRs of our MLLINERs to those of the entire galaxy, we assume that the specific star-formation rate (sSFR) is constant throughout the galaxy and therefore: SFR = SFR/M M, where the total stellar mass was taken from the MPA-JHU DR7 catalogue and is listed in Table 8 (last column), while M is the mass measured from our nuclear spectra (see Table 4). This assumption is further tested by comparing optical and FIR measurements.

SFR using STARLIGHT best fits. We followed the equation from Cid-Fernandes et al. (2013) and obtained the mean SFR surface density, by accumulating all the stellar mass formed since a look-back time of t. The mass-over-time average, is:

SFR(t) = 1/tM,

where M is the mass of stars formed at look-back time t (corresponding to M in Section 4.2). We measured three SFRs, for stellar populations younger than 10 years, for those younger than 10 years, and the total one corresponding to the entire initial mass processed into stars throughout the galaxy life (logt, column 10 in Table 4). The SFRs are listed in columns 2, 3, and 4 in Table 8. We also estimated what would be the values of STARLIGHT total SFRs when scaled to map the entire galaxy (column 5 in Table 8), as explained above.

SFR using Dn4000. We compared our results with the models obtained by Brinchmann et al. (2004), showing the relation between the sSFR and the Dn4000 index (their Figure 11). Using the nuclear M masses from the STARLIGHT fits (see Section 4.2) and our Dn4000 measurements (see Section 4.1) we obtained the mode SFRs. These values are again provided in Table 8 together with the scaled SFR if mapping the entire galaxy (columns 6 and 7).

SFR using FIR luminosity. Finally, we measured SFRs using Herschel/PACS and IRAS FIR data (see Table 1). We assumed that all the FIR luminosity is due to star formation, and that the total IR SF luminosity (TIR, the SF luminosity integrated over the range 8 - 100 m) is dominated by the FIR luminosity. Thus, LSF = L(TIR). In the case of six sources observed with Herschel, we performed the spectral energy distribution (SED) fitting to obtain L(FIR) through minimisation and using the templates of Chary & Elbaz (2001). To measure the SFR with IRAS data, we followed the same procedure applied in Tommasin et al. (2012). L is measured through F, using two IRAS bands, and following the expression provided in Sanders & Mirabel (1996):

F = 1.2610(2.58  F(60m) + F(100m)) [W m],

where F(60m) and F(100m) are the fluxes in 60m and 100m IRAS bands, respectively. As in Tommasin et al. (2012), we do not include the fluxes at 12m and 25m bands since they may be influenced by warm AGN heated dust. In the case of three MLLINERs with poor flux measurements in the 100m band, having flag quality of 1 (see table 2), we measured the total FIR flux as F = 2  F(60m) (see e.g., Rosario et al., 2012).

ID SFR1 SFR2 SFR SFR SFR SFR SFR SFR logL log(M/M) log(M/M)
F01 0.01 0.09 0.81  0.13 8.13   2.68 0.95   0.28 9.52   3.94 44.11  0.15 10.86
F02 0.18 1.24 2.61  0.41 8.12   2.67 2.5   0.21 7.77   2.34 15.78  0.05 15.15  0.45 44.9   0.24 10.88 7.45
F03 0.09 0.01 1.73  0.26 13.99  4.51 1.18   0.23 9.55   3.33 44.25  0.20 11.12
F04 0.09 0.01 2.26  0.33 17.43  5.57 1.22   0.22 9.39   3.19 44.36  0.20 11.21
F06 0.33 1.36 1.47  0.24 7.99   2.66 1.56   0.2 8.49   2.68 17.1   0.51 44.7   0.17 10.91 7.87
F07 0.21 0.48 0.67  0.12 0.82   0.28 44.22  0.30 9.86 7.14
F09 0.0 0.48 1.92  0.29 15.91  5.14 1.37   0.24 11.37  3.83 18.2   1.0 43.95  0.30 11.19
F12 0.0 2.36 0.87  0.14 9.78   3.20 0.61   0.27 6.88   3.62 20.95  1.09 44.74  0.31 10.98
F13 0.0 0.27 0.56  0.09 4.51   1.51 44.25  0.24 10.66 7.31
F14 0.29 0.37 1.13  0.18 2.71   0.91 1.1   0.22 2.64   0.94 10.42  0.55 44.0   0.20 10.41
F15 0.0 10.57 1.04  0.16 4.38   1.45 1.33   0.2 5.59   1.84 44.81  0.21 10.63 7.03
F16 0.07 0.09 0.94  0.15 4.56   1.52 2.2   0.21 10.64  3.27 11.89  0.61 44.87  0.11 10.61
F17 0.35 3.05 0.95  0.16 3.39   1.14 1.45   0.23 5.15   1.72 13.52  0.16 21.26  0.64 44.7   0.16 10.55 7.83
F19 0.0 0.49 1.66  0.24 7.25   2.35 1.55   0.24 6.77   2.22 1.25   0.04 44.33  0.12 10.82
F20 0.05 1.53 0.94  0.15 5.21   1.74 2.27   0.18 12.63  3.81 44.59  0.15 10.68
F21 0.0 1.93 1.58  0.24 1.45   0.48 1.52   0.22 1.4   0.46 44.75  0.26 10.13 7.15
F22 0.0 1.84 1.56  0.25 2.35   0.79 3.65   0.15 5.5   1.64 44.22  0.11 10.32
F23 0.0 5.18 1.19  0.2 1.93   0.65 1.42   0.21 2.31   0.77 43.24  0.44 10.25
F24 0.21 1.16 1.95  0.31 8.82   2.91 2.24   0.42 10.11  3.49 9.25   0.49 10.88
B01 0.03 0.87 0.75  0.12 3.29   1.10 1.37   0.23 5.98   2.04 44.35  0.18 10.46
B02 0.61 4.45 4.25  0.71 11.47  3.82 9.79   0.19 26.44  7,64 7.19   0.17 44.24  0.19 11.01
B03 0.36 1.06 3.22  0.49 18.94  6.14 16.47  0.06 10.33  0.52 45.36  0.10 11.22
B04 0.0 0.0 1.89  0.31 6.76   2.25 1.22   0.21 4.35   1.47 44.11  0.37 10.82 7.98
B05 0.09 0.25 4.0   0.64 29.76  9.71 3.35   0.22 24.95  7.27 15.01  0.88 44.64  0.20 11.43
B06 0.0 1.93 3.7   0.58 7.47   2.47 4.21   0.17 8.52   2.50 43.79  0.18 10.82
B07 0.54 2.17 7.57  1.13 20.91  6.74 6.81   0.24 18.81  5.41 5.31   0.09 44.71  0.25 11.26
B08 0.0 0.69 5.16  0.77 20.88  6.73 3.29   0.17 13.29  3.85 44.27  0.20 11.26 8.16
B09 0.06 0.01 0.98  0.15 5.45   1.81 43.57  0.14 10.71 7.67
B10 0.14 1.23 1.12  0.18 4.59   1.54 1.89   0.18 7.73   2.37 44.51  0.28 10.68
B11 0.0 0.68 1.64  0.28 11.66  3.89 2.66   0.12 18.93  5.51 12.02  0.63 44.75  0.12 11.06
B12 0.0 1.7 0.54  0.09 6.37   2.11 0.69   0.21 8.15   3.47 10.82 7.43
B13 0.25 3.24 1.37  0.24 7.86   2.65 14.44  0.12 82.6   23.91 18.15  0.85 45.48  0.09 10.89 8.32*
B14 0.0 1.97 5.24  0.78 17.25  5.57 2.99   0.17 9.87   2.88 44.37  0.19 11.18
B15 0.0 7.81 1.03  0.16 3.57   1.18 1.63   0.19 5.67   1.79 44.44  0.19 10.54 7.36
B16 0.0 0.0 1.28  0.2 5.14   1.69 1.33   0.22 5.35   1.79 44.35  0.21 10.71 7.93
B17 0.0 2.64 1.57  0.25 7.15   2.36 2.94   0.22 13.35  4.00 10.56  0.54 44.97  0.21 10.81
B18 0.0 5.37 1.54  0.25 2.73   0.91 1.74   0.19 3.07   0.97 43.86  0.22 10.39 7.21
B19 0.0 6.42 6.15  0.99 4.61   1.54 6.75   0.21 5.06   1.49 44.21  0.20 10.61
B20! 0.0 3.44 6.01  0.86 7.51   2.42 4.82   0.19 6.02   1.76 44.38  0.15 10.82 8.20
B21 0.0 6.95 9.25  1.41 16.77  5.44 5.89   0.19 10.68  3.08 44.72  0.30 11.17 8.03
B22! 0.0 7.84 5.29  0.86 14.64  4.83 9.68   0.19 26.78  7.71 44.6