Studying the radio continuum from nuclear activity and star formation in Giant Low Surface Brightness Galaxies
Alka Mishra, N. G. Kantharia, M. Das, D. C. Srivastava, S. N. Vogel
aa r X i v : . [ a s t r o - ph . GA ] D ec Mon. Not. R. Astron. Soc. , 000–000 (0000) Printed 8 August 2018 (MN L A TEX style file v2.2)
Studying the radio continuum from nuclear activity and starformation in Giant Low Surface Brightness Galaxies
Alka Mishra ⋆ N. G. Kantharia M. Das D. C. Srivastava S. N. Vogel Department of Physics, D.D.U. Gorakhpur University,Gorakhpur,India National Centre for Radio Astrophysics,TIFR ,Pune, India Indian Institute of Astrophysics, Koramangala,Bangalore,India Department of Astronomy, University of Maryland, College Park, MD 20742, USA
ABSTRACT
We present a multifrequency radio continuum study of seven giant low surface bright-ness (GLSB) galaxies using the Giant Metrewave Radio Telescope (GMRT). GLSB galaxiesare optically faint, dark-matter dominated systems that are poorly evolved and have large H i gas disks. Our sample consists of GLSB galaxies that show signatures of nuclear activity intheir optical spectra. We detect radio emission from the nuclei of all the seven galaxies. Fivegalaxies have nuclear spectral indices that range from 0.12 to -0.44 and appear to be core-dominated; the two galaxies have a steeper spectrum. Two of the galaxies, UGC 2936 andUGC 4422 show significant radio emission from their disks. In our 610 MHz observationsof UGC 6614, we detect radio lobes associated with the radio-loud active galactic nucleus(AGN). The lobes have a spectral index of -1.06 ± ⊙ yr − . We comparethe radio images with the near-ultraviolet (NUV) images from GALEX and near-infrared(NIR) images from 2MASS. The galaxies present a diversity of relative NUV, NIR and radioemission, supporting an episodic star formation scenario for these galaxies. Four galaxies areclassified members of groups and one is classified as isolated. Our multiwavlength study ofthis sample suggests that the environment plays an important role in the evolution of thesegalaxies. Key words: galaxies: individual- UGC 1378, UGC 1922, UGC 2936, UGC 4422, Malin 2,UGC 6614, UM 163
Low surface brightness (LSB) galaxies are late type spirals thathave properties quite distinct from regular spirals on the Hub-ble sequence. They are characterized by a central disk sur-face brightness fainter than 23 mag arcsec − in the B band(Impey & Bothun 1997; Geller et al. 2012). They have di ff use stel-lar disks (de Blok et al. 1995), large H i disks and massive dark mat-ter halos (McGaugh & de Blok 1998; van den Hoek et al. 2000).The low star formation rates (O’Neil et al. 2007) and low metal-licities (McGaugh 1994) of these galaxies suggest that they are lessevolved compared to high surface brightness galaxies (HSBGs).Chemical evolution models indicate that star formation has pro-ceeded in a stochastic manner in LSB galaxies and hence their evo-lution is slow (Bell et al. 2000; van den Hoek et al. 2000).Studies show that LSB galaxies span a wide range ofmorphologies, from the more populous dwarf LSB galax- ⋆ E-mail: [email protected] ies (Pustilnik et al. 2011) to the relatively larger giant spiralslike Malin-1 (McGaugh et al. 1995; Beijersbergen et al. 1999;Galaz et al. 2002). The larger LSB galaxies or the so-called gi-ants are relatively rare (Sprayberry et al. 1995; Bothun et al. 1990)and often isolated compared to HSBGs (Galaz et al. 2011). In fact,they are often found close to the edge of voids (Rosenbaum et al.2009). Although the disks of giant low surface brightness (GLSB)galaxies are usually poor in star formation (Auld et al. 2006);weak but distinct spiral arms are often present (McGaugh et al.1995). The barred galaxy fraction is low compared to HSBGs(Mayer & Wadsley 2004) and a significant fraction is bulgeless.The lack of disk evolution may be due to the massive dark mat-ter halos that dominate the disks of these galaxies as halos play animportant role in slowing down the formation of disk instabilities(Mihos et al. 1997). However, there are exceptions; some GLSBgalaxies show vigorous signs of disk star formation and emit copi-ous UV and H α emission (Boissier et al. 2008).GLSB galaxies have low surface brightness disks andsome have been found to host active galactic nuclei (AGN) c (cid:13) Mishra et. al.
Table 1.
Properties of the Sample GalaxiesGalaxies Type a α a (2000) δ a (2000) Optical size a inclination b V ahel ′′ a Environment Recent(h:m:s) ( ◦ : ′ : ′′ ) ( ′ × ′ ) (degree) (km s − ) = (kpc) SupernovaUGC 1378 (R)SB(rs)a 01:56:19.2 + × + × d + × + × c d Malin-2 Sd / p 10:39:52.5 + × + × d ....UM 163 SB(r)b pec 23:30:32.3 -02:27:45 1.9 × d .... a NASA Extragalactic Database b Hyperleda c Lyon Group of Galaxies (LGG) (Garcia 1993) d Low Density Contrast Extended (LDCE) groups (Crook et al. 2007;Crook et al. 2008) (Sprayberry et al. 1995; Schombert 1998). Schombert (1998) stud-ied a sample of GLSB galaxies using optical spectroscopy and con-cluded that at least 30% of the sample showed AGN activity, es-pecially those that have prominent bulges. AGN in GLSB galax-ies have also been detected in X-ray (Das et al. 2009; Naik et al.2010) with similar luminosities as of AGN in bright spirals (10 − erg s − ). In their optical study of stellar populations in bulgesof low surface brightness galaxies, Morelli et al. (2012) find themto be similar to HSBGs which they say implies that the disk-bulgeevolution is decoupled.LSB galaxies have not been extensively investigated in the ra-dio continuum. Recently, Mei et al. (2009) have conducted a sur-vey of AGNs in 196 LSB galaxies sample by Impey et al. (1996)using the spectroscopic data of SDSS DR5 and the correspondingFIRST data. They find that about 10 −
20% GLSB galaxies host anAGN as compared to the 50% found for HSBGs (Kennicutt et al.1989; Ho et al. 1997). Boissier et al. (2008) have examined the UVcolours from GALEX and the star formation e ffi ciency (SFE) ofthese galaxies. They find that the UV light extends out farther thanthe optical light and SFE are lower for some LSB galaxies. Theyalso find that the UV emission of a few LSB galaxies in the sampleresembles that observed in the XUV-disk galaxies (Gil de Paz et al.2005; Thilker et al. 2007). Besides, they also find that the FUV-NUV colour for LSB galaxies is redder than expected for star form-ing galaxies which they interpret as being due to bursts of star for-mation. In this paper we present a detailed multiwavelength (either1280 or 1420 MHz (L band), 610 MHz, 325 and 240 MHz) radiostudy of a sample of seven GLSB galaxies with the Giant Metre-wave Radio Telescope (GMRT). We study the radio morphology ofthe sources, the star forming component and the spectrum of thecentral active source in these galaxies. We refer to AGNs as a flat-spectrum sources if α > -0.5. Synchrotron self-absorption resultsin such flat spectrum. We also present a comparative study of theradio, UV (NUV) and NIR (J-band) emissions from these galaxies. See http: // irsa.ipac.caltech.edu / applications / / IM / See http: // galex.stsci.edu / GR6 / ?page = tilelist&survey = allsurveys Table 2.
Details of GALEX images used in the paperGalaxy Survey a NUV exposuretimeb
Depth b (ksec) (M AB )UGC 1378 AIS 0.1 20.5UGC 1922 NGS 1.5 23UGC 2936 GIP 3.3 ....UGC 4422 MIS 1.5 23MALIN-2 GIP 7.6 ....UGC 6614 NGS 1.5 23UM 163 GIP 3.3 .... a All sky Imaging Survey (AIS), Nearby Galaxy Survey (NGS)Medium Imaging Survey (MIS), Guest Investigator Program (GIP) b Taken from Martin et al. (2005)
In this paper, we present the radio continuum study of a sample ofseven GLSB galaxies with an aim to study the AGN. The sampleis selected from several studies in the literature (Schombert 1998;Sprayberry et al. 1995; Ramya et al. 2011) based on the followingcriteria: (1) large spirals (2) prominent bulge (3) presence of an op-tically identified AGN and (4) v sys s − . The sampleconsists of UGC 1378, UGC 1922, UGC 2936, UGC 4422, Ma-lin 2, UGC 6614 and UM 163. The basic properties of the galax-ies are listed in Table 1. The galaxy morphologies range from Sato Sd types. Four galaxies in our sample are classified as mem-bers of groups. Supernovae explosions have been recorded in threegalaxies; UGC 1922 (SN 1989s) (Mueller et al. 1989), UGC 2936(SN 1991bd) (Mueller et al. 1991) and UGC 4422 (SN 1999aa)(Armstrong & Schwartz 1999) in the recent past. The optical iden-tifications of AGNs are generally done using the spectral featuresof [NII], [SII], [OIII], H α , H β and several other emission lines.Although LSB galaxies do not contain abundant quantitiesof molecular gas, CO observations have revealed the presence ofmolecular gas in UGC 1922 (O’Neil & Schinnerer 2003) and inUGC 6614, Malin 2 (Das et al. 2006; Das et al. 2010). Four galax-ies in our sample namely UGC 1378, UGC 1922, UGC 2936and UGC 6614 have been observed in X-rays (Das et al. 2009).While a compact nuclear source has been detected in X-rays fromUGC 2936 (Das et al. 2009) and UGC 6614 (Naik et al. 2010), dif-fuse X-ray emission is detected in UGC 1378, UGC 1922 and UGC c (cid:13) , 000–000 ow Surface Brightness Galaxies Levs = 6.000E-05 * (-8, -4, 4, 8, 13)0 20 40 60 D E C L I NA T I O N ( J2000 )
610 MHz + NUV grey
RIGHT ASCENSION (J2000)01 56 30 h m s s s s s
73 18 00 o ’ "
17 45 ’ " " " "
16 45 ’ " " " "
15 45 ’ " (a)
CONT: UGC1378 IPOL 1404.571 MHZ U1378 2M1.4G.HGEOM.1158 160 162 164 D E C L I NA T I O N ( J2000 ) RIGHT ASCENSION (J2000)01 56 30 h m s s s s s
73 18 00 o ’ "
17 45 ’ " " " "
16 45 ’ " " " "
15 45 ’ " (b)
CONT: UGC1378 IPOL 616.750 MHZ U1378 2M610M.HGEOM.1158 160 162 164 D E C L I NA T I O N ( J2000 )
610 MHz + NIR grey
RIGHT ASCENSION (J2000)01 56 30 h m s s s s s
73 18 00 o ’ "
17 45 ’ " " " "
16 45 ’ " " " "
15 45 ’ " (c)
CONT: UGC1378 IPOL 312.380 MHZ U13 32CLAP2M.HGEOM.1158 160 162 164 D E C L I NA T I O N ( J2000 )
325 MHz + NIR grey
RIGHT ASCENSION (J2000)01 56 30 h m s s s s s
73 18 00 o ’ "
17 45 ’ " " " "
16 45 ’ " " " "
15 45 ’ " (d)
Figure 1. Images of UGC 1378 : The cross marks the position of the optical centre. (a) The contours showing the 610 MHz emission are plotted at 60 × (-8,-4,4,8,12,14) µ Jy beam − . The angular resolution is 7 ′′ × ′′ , PA = ◦ .82. The NUV grey scale is counts / sec. (b) The contour levels of the 1420 MHzemission are 28 × (-8,-4,4,8,13) µ Jy beam − for a beamsize of 3 ′′ × ′′ , PA = ◦ .26. The NIR grey scale are data-number. (c) Contour levels are similar to(a). (d) The contours showing the 325 MHz emission are plotted at 0.2 × (-6,-4,4,6,8,9.3) mJy beam − . The beamsize is 13 ′′ × ′′ , PA = -33 ◦ .39. J band images (wavelength = as tracer of youngmassive stellar regions (Martin et al. 2005). Table 2 gives the de-tails of the GALEX data we have used. UGC 1378 which is ob-served as part of the All sky Imaging Survey (AIS) of GALEX, hasa short exposure time of 100 s while rest of the galaxies have longerexposure times ranging from ∼ The observations were done using the GMRT (Swarup et al. 1991).The GMRT is an interferometric array of thirty antennas, each an-tenna being 45 m in diameter. Observations were carried out fromDecember 2005 to October 2012 at 240 MHz, 325 MHz, 610 MHzand L band. Flux calibration was done using scans on one of the AIPS (Astronomical Image Processing System) is distributed by the Na-tional Radio Astronomy Observatory (NRAO), which is a facility of theNational Science Foundation operated under cooperative agreement by As-sociated Universities, Inc.c (cid:13) , 000–000
Mishra et. al. standard calibrators 3C 147, 3C 286 and 3C 48, which were ob-served at the start and end of the observing runs. Phase calibrationwas done using VLA calibrator sources observed before and af-ter each scan on the target source. Details of the observations aregiven in Table 3. The raw data were converted to FITS and im-ported to AIPS . The data were reduced using standard tasks inAIPS. The calibrator data were edited and gain solutions obtained.The flux density calibrators were used for band-pass calibration.The band-pass calibrated data were averaged over a smaller numberof channels to avoid bandwidth smearing e ff ects. Wide-field imag-ing techniques were applied to the data. We generated 49 facetsat 240 MHz, 25 facets at 325, 25 facets at 610 MHz and 9 facetsat L band across the primary beam (Cornwell & Perley 1992). Datasets were phase self-calibrated up to three iterations by using strongpoint sources within the primary beam field. A final round of phaseand amplitude self-calibration was then done. Naturally (Robust =
5) and uniformly (Robust =
0) weighted (Briggs 1995) images atdi ff erent resolutions were made at all the frequencies and then cor-rected for primary beam attenuation. In order to study the compactnuclear emission we also made images after excluding the data cor-responding to large angular scales in the image i.e. short baselines. We detect radio emission from the nuclear region of all theseven galaxies at one or more radio bands. Two of the galaxies,UGC 2936 and UGC 4422, show di ff use emission which appearsto be associated with the disk. The results for the images made withRobust = = ff erences between the NVSSflux and the nuclear emission estimated from GMRT high resolu-tion images at L band. In the following subsections we discuss theresults on the individual galaxies. The spectra are shown in Figures9 (a)- (f). : Schombert (1998) identified the optical AGN in thisgalaxy. Di ff use X-ray emission is detected from the central partsof this galaxy (Das et al. 2009). The picture of environment is notclear for this galaxy.This galaxy is bright in the NIR band with emission arisingfrom the entire disk (Figures 1 (b)-(d)) but no NUV emission isdetected which is likely due to the short exposure time (Table 2).The compact radio emission from the galaxy, which is detected at325 MHz, gets resolved into two peaks with ∼ ′′ o ff set at 1420 and610 MHz. In fact, the radio peak emission at all the three frequencybands is detected to the south of the optical centre of the galaxy whereas fainter emission is detected from the optical AGN at 1420and 610 MHz. We convolved the images of the galaxy at 610 and1420 MHz to the lower resolution of the 325 MHz map to estimatethe spectrum of the emission.We detect a compact source to the east of the centre ofthe galaxy at the right ascension 01 h m s and declination + ◦ ′ ′′ .5 at 610 and 1420 MHz (Figures 1 (b) and (c)). Thisis close to a compact NIR source which is likely part of the galaxy.This source is not detected in our 325 MHz map nor is it visible inthe NVSS map. The estimated fluxes for the eastern source at 610and 1420 MHz are 0.52 ± ± α ∼ -0.9 ± UGC 1922 : AGN activity has been observed in the opticalspectrum of the galaxy by Schombert (1998). The galaxy hosteda type Ia supernova, 1989S (Mueller et al. 1989). The galaxy hasbeen detected in CO emission and the molecular gas ( ∼ × M ⊙ ) appears to be concentrated within the inner 30 ′′ of the disk(O’Neil & Schinnerer 2003). Di ff use X-ray emission has also beendetected from the central parts of this galaxy (Das et al. 2009). Thisgalaxy is classified as a member of the 37 member-group LDCE0163 (Crook et al. 2008).Radio emission is detected from the central parts of the galaxy(Figure 2 (a)). Bright NUV emission is detected along the spiralarm situated in the north-east of the centre of the galaxy and fromother compact regions. The central parts of the galaxy are bright inradio emission and NIR (Figures 2 (b)-(d)). The 610 MHz imageshows emission from the central part with an extension towards thesouth, close to the reported SN 1989S. The 1420 MHz image re-solves the emission in the central part of this low inclination galaxyinto a mini-spiral in the centre with a bright peak. Interestingly twomassive star forming complexes located north-west to the centre ofthe galaxy are detected in the NUV and NIR. No radio continuumemission is detected in any of the observed radio bands. UGC 2936 : This galaxy is almost edge-on in morphology.The AGN in this GLSB galaxy was identified by Sprayberry et al.(1995). The AGN is detected in X-ray with a luminosity of 1.8 × ergs s − (Das et al. 2009). A type II supernova, SN 1991bdhas been recorded in this galaxy (Mueller et al. 1991). Unlikemost GLSB galaxies with bulges, it has a boxy / peanut shapedbulge rather than a classical bulge. This suggests that the galaxymay be undergoing secular evolution of its bar into a boxy bulge(Raha et al. 1991).The galaxy is observed at 1280 and 610 MHz with the GMRT.Faint NUV emission is detectable from the centre of the galaxy(Figure 3 (a)). The entire disk of the galaxy is detected in NIR andradio bands in addition to the intense emission from the active nu-cleus (Figures 3 (b) and (c)). Good correlation is seen between theextent of the NIR and radio disks due to star formation. UGC 4422 : The AGN in this barred galaxy was identi-fied by Schombert (1998). No X-ray emission has been detectedfrom the centre of this galaxy with an upper limit of 10 erg-s − (Das et al. 2009). This galaxy hosted a type Ia supernovaSN 1999aa (Armstrong & Schwartz 1999). It is a member of thegroup LGG 159 (Garcia 1993) and LDCE 571 with 22 members(Crook et al. 2008).The 610 MHz emission is confined to the central parts of thegalaxy while bright NUV emission is observed along the spiralarms and to the west of the centre are also seen (Figure 4 (a)). Theemission at 1420 MHz and 610 MHz is concentrated in the centralparts of the galaxy while the 325 MHz emission extends along thespiral arms (Figures 4 (b)- (d)). The 610 MHz image shows emis- c (cid:13) , 000–000 ow Surface Brightness Galaxies Table 3.
Details of GMRT ObservationsGalaxies Obs. Date Band Calibrators BW c t source N Ant . d (dd / mm / yy) (MHz) Phase S phase (Jy) a Flux S flux (Jy) b (MHz) (hrs)UGC 1378 26 / /
12 325 0114 +
483 5.9 3C48 45.0 32 3.5 2830 / /
06 610 0217 +
738 2.1 3C48, 3C147 29.4, 38.2 16 6.0 2730 / /
11 1420 0217 +
738 1.8 3C48 16.8 16 5.5 29UGC 1922 27 / /
12 325 3C48 44.5 3C48 44.5 32 3.5 2831 / /
05 610 3C48 29.3 3C48 29.3 16 6.0 2831 / /
11 1420 3C48 16.4 3C48 16.4 16 6.0 29UGC 2936 14 / /
07 610 0323 +
055 5.3 3C147 38.2 16 3.0 2819 / /
07 1280 0323 +
055 3.4 3C147 23.6 16 3.0 28UGC 4422 26 / /
11 325 0909 +
428 17.6 3C147, 3C286 54.0,26.4 32 2.0 2814 / /
07 610 0735 +
331 5.3 3C147 38.2 16 2.5 2729 / /
11 1420 0842 +
185 1.1 3C286, 3C147 22.2, 14.8 16 5.5 29Malin-2 09 / /
06 240 3C241 10.1 3C147, 3C286 48.6, 25.8 16 5.5 2809 / /
06 610 3C241 4.3 3C147, 3C286 36.4, 20.6 16 5.5 2811 / /
11 1280 3C241 2.0 3C147 23.7 32 1.0 27UGC 6614 31 / /
05 240 1123 +
055 6.4 3C48, 3C286 47.3, 25.8 16 6.5 2731 / /
05 610 1123 +
055 3.4 3C48, 3C286 29.3, 20.7 16 6.5 2712 / /
11 1280 1120 +
143 2.7 3C147 23.5 32 1.45 29UM 163 26 / /
11 325 2225-049 13.9 3C48, 3C147 42.8, 46.7 32 2.5 2901 / /
06 610 2225-049 10.6 3C48 29.4 16 4.0 2828 / /
06 1420 2225-049 8.4 3C48, 3C286 16.2,15.0 04 5.0 27 a Flux densities (S phase ) of phase calibrators from GETJY, task in AIPS b Flux densities (S flux ) of flux calibrators from SETJY, task in AIPS c Observing Bandwidth d Average number of antennas working during the observations
Table 4.
Results of GMRT ObservationsGalaxy Band Low resolution (Robust = a High resolution (Robust = a StructureFlux Density b Beam Size P.A. σ Flux Density b Beam Size P.A. σ MHz mJy ( ′′ × ′′ ) degree mJy / b mJy ( ′′ × ′′ ) degree mJy / bUGC 1378 325 2.2 ± ×
18 60.4 0.43 1.7 ± ×
11 -33.3 0.18 Compact core610 1.7 ± × ± × ± × ± × ± ×
16 57.2 1.44 68.1 ± × ± ×
12 -82.8 0.20 56.0 ± × + extended1420 35.1 ± × ± × ± ×
14 -42.3 0.28 46.4 ± × + extended1280 29.3 ± × c × + extendedUGC 4422 325 49.9 ± ×
18 -11.6 0.52 42.8 ± × + extended610 10.1 ± × ± × ± × ± × ± ×
30 -23.3 0.76 5.2 ± ×
14 -31.9 0.94 Compact core610 5.0 ± ×
19 33.5 0.18 5.0 ± × ± × ± × < d ×
36 -4.6 1.71 < d ×
14 -64.7 1.15 Not detected610 10.8 ± ×
16 -50.2 0.16 10.2 ± × + extended1280 4.4 ± × ± × ± ×
11 61.8 0.27 26.1 ± × ± ×
10 -8.36 0.23 13.5 ± × + extended1420 8.8 ± × ± × a See Briggs (1995) b A systematic flux density error of 15% of total flux at 240, 325 and 610 MHz and 5% at L band, is assumed. c Galaxy emission is fragmented so the flux density is not measured. d UGC 6614 is not detected at 240 MHz and 4 σ limit is tabulated.c (cid:13) , 000–000 Mishra et. al. D E C L I NA T I O N ( J2000 )
610 MHz + NUV grey
RIGHT ASCENSION (J2000)02 27 52 h m s s s s s s s
28 14 00 o ’ "
13 30 ’ " "
12 30 ’ " "
11 30 ’ " " (a) CONT: UGC1922 IPOL 1369.717 MHZ U1922 2M1.4M.HGEOM.184 86 88 90 D E C L I NA T I O N ( J2000 ) RIGHT ASCENSION (J2000)02 27 49 h m s s s s s s s s
28 13 15 o ’ " "
12 45 ’ " " " "
11 45 ’ " (b)
CONT: UGC1922 IPOL 607.125 MHZ U1922 2M610M.HGEOM.184 86 88 90 D E C L I NA T I O N ( J2000 )
610 MHz + NIR grey
RIGHT ASCENSION (J2000)02 27 49 h m s s s s s s s s
28 13 15 o ’ " "
12 45 ’ " " " "
11 45 ’ " (c)
CONT: UGC1922 IPOL 313.943 MHZ U1922 2M325M.HGEOM.284 86 88 90 D E C L I NA T I O N ( J2000 )
325 MHz + NIR grey
RIGHT ASCENSION (J2000)02 27 49 h m s s s s s s s s
28 13 15 o ’ " "
12 45 ’ " " " "
11 45 ’ " (d)
Figure 2. Images of UGC 1922 : The cross marks the position of optical centre and the triangle at the south-east marks the position of the supernova, SN1989S. (a) The contours showing the 610 MHz emission are plotted at 0.1 × (-8,-4,4,8,16,32,64,128,256,330) mJy beam − . The angular resolution is 7 ′′ × ′′ ,PA = ◦ .11. The NUV grey scale is counts / sec. (b) The contour levels of the 1420 MHz emission are 60 × (-8,-4,4,8,16,64,256,350) µ Jy beam − for beamsizeof 3 ′′ × ′′ , PA = ◦ .55. The NIR grey scale are data-number. (c) Contour levels are similar to (a). (d) The contours showing the 325 MHz emission areplotted at 0.8 × (-8,-4,4,8,16,32,64,75) mJy beam − . The beamsize is 12 ′′ × ′′ , PA = ◦ .68. sion at the onset of the spiral arms in the north-west and south-east directions. Interestingly, the NIR emission from this galaxy isextended. The di ff use extended emission seen at 325 MHz alongthe spiral arms is not detected at 610 or 1420 MHz. The di ff useemission at 325 MHz in the southern arm has a brightness of 1.1mJy / beam (see Figure 4 (d)). Using the 4 σ limit of our low res-olution map at 610 MHz of 0.48 mJy / beam; we estimate that thespectral index is < − . ff use emission. We made mapsafter excluding the shorter baselines to remove extended emissionbut did not detect an unresolved source in any of the wavebands.The optical image of this galaxy shows a circumnuclear ring-like feature at the centre which is also discernible in our 1420 MHzimage. To confirm its presence in our 1420 MHz image, we tooka one-dimensional cut along right ascension (see Figure 5). Thisshows the presence of three peaks; the outer ones likely define a ring surrounding the central peak which is likely the optical AGN.We measure the size of the ring ∼ ′′ which at the distance of UGC4422 corresponds to a diameter of 1.8 kpc. Comer´on et al. (2010)detect a star forming circumnuclear ring in their HST optical imageof the galaxy. They estimate the semimajor axis of the ring to be5.4 ′′ which corresponds to 1.6 kpc. A type Ia supernova, SN 1999aawas reported in the northern part of the galaxy but we do not detectany radio emission from that region. Malin 2 : The nucleus of the galaxy shows AGN activity at op-tical wavelengths (e.g. Ramya et al. 2011). It has been detected inCO emission and the molecular gas ( ∼ × M ⊙ ) is found to ex-tend towards the west in the galaxy disk (Das et al. 2006; Das et al.2010). It is not detected in the continuum at millimetre wavelengths(Das et al. 2006). This galaxy is classified as an isolated galaxy.The central AGN is detected at 610 MHz with no radio emis- c (cid:13) , 000–000 ow Surface Brightness Galaxies CONT: UGC2936 IPOL 614.875 MHZ UGC2936.NUV.HGEOM.120 40 60 80 D E C L I NA T I O N ( J2000 )
610 MHz + NUV grey
RIGHT ASCENSION (J2000)04 02 51 h m s s s s s s s
01 58 45 o ’ " " " "
57 45 ’ " " " " (a) CONT: UGC2936 IPOL 1286.750 MHZ U2936 2M1.2M.HGEOM.2155 160 165 170 D E C L I NA T I O N ( J2000 ) RIGHT ASCENSION (J2000)04 02 51 h m s s s s s s s
01 58 45 o ’ " " " "
57 45 ’ " " " " (b) CONT: UGC2936 IPOL 614.875 MHZ U2936 2M610M.HGEOM.1155 160 165 170 D E C L I NA T I O N ( J2000 )
610 MHz + NIR grey
RIGHT ASCENSION (J2000)04 02 51 h m s s s s s s s
01 58 45 o ’ " " " "
57 45 ’ " " " " (c) Figure 3. Images of UGC 2936 : The cross marks the optical centre of the galaxy and the triangle in the north-east marks the position of the supernova, SN1991bd. (a) The contours showing the 610 MHz emission are plotted at 0.3 × (-8,-4,4,8,16,32,64,66) mJy beam − . The angular resolution is 15 ′′ × ′′ , PA = ◦ .48. The NUV grey scale is counts / sec. (b) The contour levels of the 1280 MHz emission are 76 × (-8,-4,4,8,16,32,64,75) µ Jy beam − for beamsize of 3 ′′ × ′′ , PA = ◦ .44. The NIR grey scale are data-number. (c) Contour levels are similar to (a). M i c r o J Y / B E A M ARC SEC-4 -3 -2 -1 0 1 2 3 4350300250200150
Figure 5.
A slice along the right ascension passing through the centre of thegalaxy in the 1420 MHz image of UGC 4422 which shows the presence ofa central peak and a circumnuclear ring. sion associated with the intense star forming disk that is detectedin the NUV (Figure 6 (a)). Intense NUV emission indicates a re-cent burst of star formation that has been triggered in the galaxy.The south-western spiral arm shows a particularly vigorous burst ofstar formation. Unresolved emission from the centre of the galaxyis detected at all the three observed frequencies i.e. 1280 MHz,610 MHz and 240 MHz (Figures 6 (b)-(d)). The NUV emission isintense and defines the spiral arms of the galaxy whereas the NIRemission is featureless with only an intense core being detected. Asingle power law fit to the three radio points resulted in a spectralindex of -0.23 ± ± ± UGC 6614 : This is a relatively well studied GLSB galaxy withthe AGN detected in mm-wave continuum (Das et al. 2006), optical(Schombert 1998; Ramya et al. 2011), X-ray (Naik et al. 2010) andNIR (Rahman et al. 2007) wavelengths. It has a low inclination anda large, prominent bulge surrounded by a faint ring-like structure(Rahman et al. 2007). This ring is clearly seen in the Spitzer NIRimage (Hinz et al. 2007; Rahman et al. 2007) and simulations sug-gest that it is the result of galaxy collision in the past (Mapelli et al.2008). The radio spectrum of the core between 100 GHz and 1.4GHz is found to be flat (Das et al. 2006). It has been detected inCO emission. The emission is detected o ff set from the galaxy cen-tre and the molecular gas mass is ∼ × M ⊙ (Das et al. 2006).This galaxy is a member of the group LDCE 829 (Crook et al. 2007,2008) which has three members.Like Malin 2, NUV emission is detected over the entire diskwith star formation seen in the ring and along the spiral arms ofthe galaxy (Figure 7 (a)). Radio emission at 610 MHz is mostlyconfined to the central region - from the AGN and extended alongwhat appears to be jets / lobes along the north-west and south-eastdirection. No NUV emission is coincident with these features. Ra-dio emission is detected from several compact regions near thecentral ring and along the spiral arms at 610 MHz. The emissionis coincident with the NUV peaks indicating that its origin is starformation. Figures 7 (b) and (c) show the zoomed-in contour im-ages of the radio continuum emission from the galaxy at 1280 MHzand 610 MHz superposed on the 2MASS NIR image in grey scale.The AGN core is detected at 1280 MHz and NIR J band but noemission is detected from the radio jets / lobes. We detect a com-pact radio source to the north-west of the central core near rightascension 11 h m s .75 and declination 17 ◦ ′ ′′ at 610 MHz co-incident with NUV emission. The estimated peak brightness of thissource at 610 MHz is 0.9 ± / b. Using the 4 σ limit at 1280MHz of 0.24 mJy implies a spectral index steeper than − . c (cid:13) , 000–000 Mishra et. al.
CONT: UGC4422 IPOL 614.250 MHZ U4422 NUV610.HGEOM.10 10 20 30 D E C L I NA T I O N ( J2000 )
610 MHz + NUV grey
RIGHT ASCENSION (J2000)08 27 48 h m s s s s s s s s
21 30 00 o ’ "
29 30 ’ " "
28 30 ’ " "
27 30 ’ " (a)
CONT: UGC4422 IPOL 1400.122 MHZ U4422 2M1.4G.HGEOM.1119 120 121 122 123 D E C L I NA T I O N ( J2000 ) RIGHT ASCENSION (J2000)08 27 46 h m s s s s s s s s s s
21 29 45 o ’ " " " "
28 45 ’ " " " "
27 45 ’ " (b)
CONT: UGC4422 IPOL 614.250 MHZ U4422 2M610M.HGEOM.3119 120 121 122 123 D E C L I NA T I O N ( J2000 )
610 MHz + NIR grey
RIGHT ASCENSION (J2000)08 27 46 h m s s s s s s s s s s
21 29 45 o ’ " " " "
28 45 ’ " " " "
27 45 ’ " (c)
CONT: UGC4422 IPOL 313.943 MHZ U4422 2M325M.HGEOM.6119 120 121 122 123 D E C L I NA T I O N ( J2000 )
325 MHz + NIR grey
RIGHT ASCENSION (J2000)08 27 46 h m s s s s s s s s s s
21 29 45 o ’ " " " "
28 45 ’ " " " "
27 45 ’ " (d)
Figure 4. Images of UGC 4422 : The cross marks the optical centre of the galaxy and triangle at the north marks the position of the supernova, SN 1999aa. (a)The contours showing the 610 MHz emission are plotted at 0.1 × (-6,-4,4,6,8,10,12) mJy beam − . The angular resolution is 6 ′′ × ′′ , PA = -35 ◦ .49. The NUVgrey scale is counts / sec. (b) The contour levels of the 1420 MHz emission are 28 × (-6,-4,4,6,8,10,13) µ Jy beam − for beamsize 3 ′′ × ′′ , PA = ◦ .05. TheNIR grey scale are data-number. (c) Contour levels are similar to (a). (d) The contours showing the 325 MHz emission are plotted at 0.4 × (-4,-3,3,4,8,10,12,14)mJy beam − . The beamsize is 10 ′′ × ′′ , PA = ◦ .53. the galaxy and might be a background radio source which shows achance coincidence. With the current data, we are unable to com-ment on this region any further.We estimate the peak flux density of the core to be 4.0 ± α ∼ + ± σ limit of 4.6 mJy and this is consistentwith the estimated α . We also estimated the spectral index of thelobe emission as follows. We subtracted the core flux density at1280 MHz of 4.4 mJy from the NVSS flux density of 7.5 mJy andassumed the remaining flux density as arising in the jets / lobes. Incombination with the emission at 610 MHz obtained after subtract-ing the core emission from the total emission at 610 MHz we obtain the spectral index, α ∼ − . ± .
12 between 610 and 1280 MHzfor the jets / lobes. UM 163 : This galaxy, also known as 2327-0244, was firststudied by Sprayberry et al. (1995) and was found to host an AGN.The galaxy has a strong bar, a prominent bulge and two trailing spi-ral arms. Its morphology is similar to early type spirals but its diskis low surface brightness in nature.This galaxy is detected at all the observed frequencies. NUVemission is detected from the centre of the galaxy; from a ring closeto the centre and along the spiral arms (Figure 8 (a)). The 610 MHzradio emission is confined mainly to the core of the galaxy witha 4 σ extension seen towards the south. The radio emission is re-solved at all the wavebands. The bar is prominent in the NIR image(Figures 8 (b)-(d)). At 1420 MHz, the emission is extended towardsthe north-west. Our low resolution image at 325 MHz shows the c (cid:13) , 000–000 ow Surface Brightness Galaxies CONT: MALIN2 IPOL 607.000 MHZ MAL2 610NUV.HGEOM.10 5 10 D E C L I NA T I O N ( J2000 )
610 MHz + NUV grey
RIGHT ASCENSION (J2000)10 39 56 h m s s s s s s s s s
20 51 45 o ’ " " " "
50 45 ’ " " " "
49 45 ’ " (a)
CONT: MALIN2 IPOL 1284.698 MHZ MALI2 2M1.2M.HGEOM.178 80 82 84 D E C L I NA T I O N ( J2000 ) RIGHT ASCENSION (J2000)10 39 54.0 h m s s s s s s s
20 51 00 o ’ "
50 45 ’ " " " (b) CONT: MALIN2 IPOL 607.000 MHZ MALI2 2M610M.HGEOM.178 80 82 84 D E C L I NA T I O N ( J2000 )
610 MHz + NIR grey
RIGHT ASCENSION (J2000)10 39 54.0 h m s s s s s s s
20 51 00 o ’ "
50 45 ’ " " " (c) CONT: MALIN2 IPOL 240.375 MHZ MALI2 2M240M.HGEOM.478 80 82 84 D E C L I NA T I O N ( J2000 )
240 MHz + NIR grey
RIGHT ASCENSION (J2000)10 39 54.0 h m s s s s s s s
20 51 00 o ’ "
50 45 ’ " " " (d) Figure 6. Images of Malin 2 : The cross marks the optical centre of the galaxy. (a) The contours showing the 610 MHz emission are plotted at 0.2 × (-8,-4,4,8,16,28) mJy beam − . The angular resolution is 7 ′′ × ′′ , PA = ◦ .69. The NUV grey scale is counts / sec. (b) The contour levels of the 1280 MHz emissionare 89 × (-8,-4,4,8,16,32) µ Jy beam − for beamsize is 3 ′′ × ′′ , PA = -76 ◦ .85. The NIR grey scale are data-number. (c) Contour levels are similar to (a). (d)The contours showing the 240 MHz emission are plotted at 0.8 × (-6,-4,4,6,7.5) mJy beam − . The beamsize is 37 ′′ × ′′ , PA = -31 ◦ .92. emission arising along the bar and also extended perpendicular tothe bar. The emission is likely to be due to the AGN core and jets.The 610 MHz and the 1420 MHz maps are convolved to the reso-lution of the 325 MHz image to estimate the spectrum of emission.We estimate α ∼ -0.82 ± Our objective of the present investigation is to study the low radiofrequency spectrum of the central AGN, star forming disk and theinfluence of the environment on our sample of GLSB galaxies usingour high resolution, high sensitivity continuum images at 240, 325,610 MHz and L band.
We detect radio continuum emission from the centre of all sevengalaxies in our sample. Our high resolution data allows us to deter-mine the core spectrum. We distinguish the spectrum as steep if α< -0.5 or flat if α > -0.5; S ∝ ν α . In AGNs, compact flat-spectrumnuclear radio cores are widely accepted as the signature of syn-chrotron self-absorption. The radio spectra of the nuclear sourcesof the galaxies, UGC 1922, UGC 2936, UGC 4422, Malin 2 andUGC 6614 exhibit spectral indices ranging from α = + .
12 to − .
44. We interpret these flatter spectra as arising from the radiocore. In case of UGC 1378 and UM 163 the spectrum is steeper.We note that in UGC 1378, the peak radio emission is displaced tothe south of the optical centre indicating an origin distinct from theAGN. Only faint radio emission is detected from the optical centrewhich is likely to be due to the radio core. However, we are not able c (cid:13) , 000–000 Mishra et. al.
CONT: UGC6614 IPOL 609.438 MHZ U6614NUV 610.HGEOM.10 5 10 D E C L I NA T I O N ( J2000 )
610 MHz + NUV grey
RIGHT ASCENSION (J2000)11 39 22 h m s s s s s s s s
17 11 00 o ’ "
10 30 ’ " "
09 30 ’ " "
08 30 ’ " "
07 30 ’ " " (a) CONT: UGC6614 IPOL 1268.812 MHZ U6614 2M1.2M.HGEOM.170 75 80 D E C L I NA T I O N ( J2000 ) RIGHT ASCENSION (J2000)11 39 16.5 h m s s s s s s s s
17 09 15 o ’ " "
08 45 ’ " " (b) CONT: UGC6614 IPOL 609.438 MHZ U6614 2M610.HGEOM.170 75 80 D E C L I NA T I O N ( J2000 )
610 MHz + NIR grey
RIGHT ASCENSION (J2000)11 39 16.5 h m s s s s s s s s
17 09 15 o ’ " "
08 45 ’ " " (c) Figure 7. Images of UGC 6614 : The cross marks the optical centre of the galaxy. (a) The contours showing the 610 MHz emission are plotted at 0.1 × (-6,-4,4,6,8,12,24,39) mJy beam − . The angular resolution is 8 ′′ × ′′ , PA = -85 ◦ .92. The NUV grey scale is counts / sec. (b) The contour levels of the 1280MHz emission are 60 × (-8,-4,4,8,16,32,64,73) µ Jy beam − for beamsize 3 ′′ × ′′ , PA = ◦ .26. The NIR grey scale are data-number. (c) Contour levels aresimilar to (a). Table 5.
The spectrum of the emission from the sample galaxies between 240 MHz and 1420 MHz along with nuclear flux densities due to AGN and disk emissionGalaxy Frequency Nuclear Flux α ( S ∝ ν α ) Integrated α NVSS Flux b (NVSS - NuclearDensity (Nucleus) Flux Density a (Integrated) Density Flux Density) c MHz mJy mJy mJy mJyUGC 1378 325 1.7 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± α ∼ -0.44 ± ± α ∼ -0.38 ± ± ± ± ± ± ± d α ∼ -2.37 ± ± ± ± α ∼ -1.05 ± ± ± ± ± ± ± ± ± ± ± ± ± ± < < ± α ∼ + ± ± α ∼ -1.13 ± e ± ± ± ± ± ± ± ± ± ± ± ± ± ± a Robust = b NVSS Flux density estimated from the NVSS images c Likely to be due to the star forming disk and used in Table 6 d The 325 MHz detects di ff use emission from the star forming disk which is not detected at the other frequencies, results in a steep spectrum of α = -2.37 ± e Jet / lobe emission included c (cid:13)000
The spectrum of the emission from the sample galaxies between 240 MHz and 1420 MHz along with nuclear flux densities due to AGN and disk emissionGalaxy Frequency Nuclear Flux α ( S ∝ ν α ) Integrated α NVSS Flux b (NVSS - NuclearDensity (Nucleus) Flux Density a (Integrated) Density Flux Density) c MHz mJy mJy mJy mJyUGC 1378 325 1.7 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± α ∼ -0.44 ± ± α ∼ -0.38 ± ± ± ± ± ± ± d α ∼ -2.37 ± ± ± ± α ∼ -1.05 ± ± ± ± ± ± ± ± ± ± ± ± ± ± < < ± α ∼ + ± ± α ∼ -1.13 ± e ± ± ± ± ± ± ± ± ± ± ± ± ± ± a Robust = b NVSS Flux density estimated from the NVSS images c Likely to be due to the star forming disk and used in Table 6 d The 325 MHz detects di ff use emission from the star forming disk which is not detected at the other frequencies, results in a steep spectrum of α = -2.37 ± e Jet / lobe emission included c (cid:13)000 , 000–000 ow Surface Brightness Galaxies to estimate the spectral index of this source separately. In case ofUM 163, it is likely that the radio emission includes contributionfrom the jets / lobes thus leading to its steep spectrum. We note thatthe flux density measured by NVSS is similar to our high resolutionL band image (see Figure 9 (f)).Recent studies show that only about 15 - 20% of normal LSBgalaxies host an AGN (Burkholder et al. 2001, Mei et al. 2009)and most appear to be associated with large bulges (Schombert1998). The nuclear black hole masses of GLSB galaxies are rel-atively low and lie in the range 10 − M ⊙ (Ramya et al. 2011).GLSB galaxies lie below the M BH − σ relation followed by othergalaxies (Ramya et al. 2011). Thus though these galaxies are verymassive, their AGN are less evolved than brighter galaxies on the M − σ correlation. This is possibly because the dark matter halosin these galaxies inhibit the formation of disk instabilities that trig-ger star formation and lead to gas infall into their nuclear regions(Mayer & Wadsley 2004). The slower gas infall rate probably re-sults in a lower AGN fueling rate, which leads to lower nuclearblack hole masses.Thus, on the basis of an analysis of low frequency radio dataof a sample of seven GLSB galaxies with optically identified AGN,we find that five out of them have radio emission associated withAGN core having spectral indices > -0.44, thereby exhibiting sig-nature of synchrotron self-absorption. Two galaxies of our sample,UGC 6614 and UM 163, appear to have additional emission fromjets / lobes. However, higher resolution images are required to disen-tangle the contributions from the core of the AGN and the jets / lobesin these two cases. Radio jets are often considered to be tracers of AGN activity in themore massive radio galaxies even if the core is not detected. How-ever, radio jets in spiral galaxies are relatively rare probably be-cause their AGN are relatively weaker than those found in the moremassive radio galaxies. Gallimore et al. (2006) studied a sample of43 Seyfert galaxies and found that 19 (44 %) showed extended ra-dio jets. In our sample of seven GLSBs, we clearly detect a jet inUGC 6614 and a jet-like feature is detected in UM 163. Thus wedetect jet-like features in 2 out of 7 galaxies i.e. about 28%. How-ever the sample size needs to be increased to make a valid com-parison. The Seyfert nucleus of the spiral NGC 4258 is one of rareexamples of extended jets in spiral galaxies (Krause et al. 2007) asis NGC 1275 (Perseus A). The jet length scales in UGC 6614 andUM 163 are 6.8 kpc and 13 kpc respectively and both jets end wellwithin the inner optical disk, which is similar to what is seen inother spirals (Laine & Beck 2008). The spectrum of emission fromthe jets in UGC 6614 has a spectral index of -1.06 ± All the seven galaxies in our sample have GALEX UV and 2MASSNIR images. Surprisingly, five of the galaxies, UGC 1922, UGC4422, Malin 2, UGC 6614 and UM 163 show bright NUV emis-sion over their entire disk which could be due to a recent burstof massive star formation in the galaxy. All the galaxies show ra-dio emission associated with the nuclear region, additionally UGC2936 and UGC 4422 show radio emission associated with the starforming disk. The NVSS flux densities at 1.4 GHz are higher than what we record in our high resolution L band images for many ofthe galaxies (see Table 5). We infer that the excess radio emissionin most cases is from the disk of the galaxies. In particular, we findthat more than half of the emission recorded by NVSS at 1.4 GHzis associated with a di ff use, disk component for UGC 4422, UGC1378 and UGC 2936 (Table 5). On the other hand, 80% of the totalemission at 1.4 GHz from UGC 1922 and UM 163 and 50 % emis-sion from Malin 2 and UGC 6614 arise in the active nucleus. Allthese galaxies show bright extended NUV disks indicating a recentburst ( ∼
100 million yrs) of star formation. This starburst is yet toresult in su ffi cient number of supernova remnants required for thenon-thermal radio emission to be detectable. The synchrotron life-time is expected to be 100 Myr at 1.5 GHz if strength of magneticfield (B) is 5 µ G (Condon 1992). Moreover in case of UGC 4422,NUV, NIR and radio emission are all detected from the disk indicat-ing continuous star formation in the galaxy. It would be interestingto derive estimates of the stellar ages in these galaxies.We detect di ff use radio emission from the disk of UGC 2936.Faint NUV is detected from the centre of the galaxy. The presenceof NIR emission and absence of NUV from the disk of the galaxysuggests an older burst of star formation, probably more than fewGyr ago. The star formation episode has since been quenched andthe galaxy is not prominent in the NUV. No extended radio emis-sion is detected from the disks of UGC 1922, Malin 2, UGC 6614and UM 163.The combination of our radio results with the NIR and NUVresults seem to give support to the episodic star formation in oursample GLSB galaxies. It is not clear what leads to a starburstepisode or what leads to it being quenched but the range of fea-tures displayed by our sample of GLSB galaxies suggests that thesegalaxies are intriguing and might likely be an evolutionary step formany ’normal’ galaxies. GLSB evolve at a much slower rate be-cause of the e ff ect of the dominant dark matter halos. LSB galaxies generally have a lower SFR ( ∼ . ⊙ / yr; McGaugh1994) and lower metallicity than HSBGs. High resolution imag-ing of star forming regions in LSB galaxies show that their diskscontain HII regions that are very similar to those found in brightergalaxies (Schombert et al. 2013). Their disks are gas rich but thesurface density is below the critical threshold for triggering star for-mation (van der Hulst et al. 1993; Pickering et al. 1997; Das et al.2010). We estimated the SFR in the sample of our galaxies usingthe relation as given in Condon (1992) for M > ⊙ and the resultsare reported in Table 6. For this, we subtracted the flux density ofthe nuclear source as obtained from our high resolution images at Lband from the flux density of the galaxy, estimated from the lowerresolution NVSS image. We find that for UGC 1922 and UGC2936, the flux densities estimated from our low resolution GMRTmaps are similar to the NVSS values. In UM 163, our low reso-lution GMRT map gives a larger flux density than the NVSS mapand for the remaining galaxies the NVSS flux densities are larger.We assume that the excess NVSS emission is due to disk star for-mation, and non-thermal in nature. Using above flux densities andassuming the non-thermal spectral index α = − .
8, we estimatedaverage star formation rates listed in Table 6. The radio emissioncan be used as a direct probe of the very recent star-forming activ-ity in normal and starburst galaxies. Nearly all of the radio emis-sion at lower frequencies from such galaxies is synchrotron radi-ation from relativistic electrons and free-free emission from H IIregions. Massive stars (M > ⊙ ) produce the supernovae whose c (cid:13) , 000–000 Mishra et. al.
CONT: 2327-024 IPOL 616.875 MHZ UM163NUV 610.HGEOM.10 5 10 15 20 D E C L I NA T I O N ( J2000 )
610 MHz + NUV grey
RIGHT ASCENSION (J2000)23 30 36 h m s s s s s -02 26 30 o ’ " "
27 00 ’ " " " "
28 00 ’ " " " " (a) CONT: 2327-024 IPOL 1374.311 MHZ U163 1.4.2MA.HGEOM.1128 130 132 134 D E C L I NA T I O N ( J2000 ) RIGHT ASCENSION (J2000)23 30 34.5 h m s s s s s s s s s s -02 27 15 o ’ " " "
28 00 ’ " " (b) CONT: 2327-024 IPOL 616.875 MHZ U163 610.2MA.HGEOM.1128 130 132 134 D E C L I NA T I O N ( J2000 )
610 MHz + NIR grey
RIGHT ASCENSION (J2000)23 30 34.5 h m s s s s s s s s s s -02 27 15 o ’ " " "
28 00 ’ " " (c) CONT: 2327-024 IPOL 314.464 MHZ U163 325.2MA.HGEOM.1128 130 132 134 D E C L I NA T I O N ( J2000 )
325 MHz + NIR grey
RIGHT ASCENSION (J2000)23 30 34.5 h m s s s s s s s s s s -02 27 15 o ’ " " "
28 00 ’ " " (d) Figure 8. Images of UM 163 : The cross marks the position of the optical centre. (a) The contours showing the 610 MHz emission are plotted at 0.2 × (-8,-4,4,8,16,25,28) mJy beam − . The angular resolution is 7 ′′ × ′′ , PA = ◦ .41. The NUV grey scale is counts / sec. (b) The contour levels of the 1420 MHzemission are 0.2 × (-8,-4,4,8,14) mJy beam − for beamsize 5 ′′ × ′′ , PA = ◦ .06. The NIR grey scale are data-number. (c) Contour levels are similar to (a).(d) The contours showing the 325 MHz emission are plotted at 0.4 × (-8,-4,4,8,16,32,50) mJy beam − . The beamsize is 12 ′′ × ′′ , PA = ◦ .70. remnants (SNRs) accelerate the relativistic electrons and are usedas a tracer of star formation. It may be noted from Table 6 that theSFRs of the sample galaxies range from 0.15 to 3.6 M ⊙ / yr withfive galaxies showing SFRs less than 1 M ⊙ / yr. For comparison, theSFRs that we obtained from literature which were estimated usingeither the UV or total infrared emission are also listed there. Theserange from 0.8 to 4.3 M ⊙ / yr. This suggests that the SFRs of thesegalaxies show a wide range similar to the HSBGs and are likely fordi ff erent epochs.Boissier et al. (2008) have examined the UV properties of asample of 13 GLSB galaxies which include three galaxies from oursample; UGC 2936, Malin 2 and UM 163. Using GALEX data theyfind that the UV disks of 9 out of their sample of 13 galaxies areextended and the FUV-NUV colours of these galaxies are redderthan normal galaxies which they interpret as being due to episodic bursts of star formation. Four galaxies in our sample show brightNUV disk.We estimated supernova rates which range from 0.006 to 0.15yr − (see Table 6). For comparison the supernova rate in M82 is0 . yr − (Condon 1992). Two of the galaxies in our sample, namelyUGC 1922 and UGC 4422 have hosted a type Ia supernova -SN1989s and SN1999aa respectively. Since the progenitor systemof these type of supernovae are believed to consist of a binary with awhite dwarf member, it can indicate star formation which was trig-gered more than a few Gyrs ago in these galaxies with the low massstars evolving into type Ia supernovae. We note that NUV, NIR andradio continuum emission from the stellar disk are detected in thecase of UGC 4422 which could indicate continuous star formation.In case of UGC 1922, most of the resolved radio continuum arisesclose to the centre of the galaxy. NUV is detected from arcs resem-bling spiral arms around the centre of the galaxy whereas di ff use c (cid:13) , 000–000 ow Surface Brightness Galaxies F L U X ( m Jy ) FREQUENCY (MHz)UGC 1378 α = −0.86+/−0.12 (a)
100 100 1000 10000 F L U X ( m Jy ) FREQUENCY (MHz)UGC 1922 α = −0.39+/−0.09 (b)
10 100 1000 10000 F L U X ( m Jy ) FREQUENCY (MHz)UGC 4422 α = −0.41+/−0.07 (c)
10 100 1000 10000 F L U X ( m Jy ) FREQUENCY (MHz)Malin 2 α = −0.23+/−0.11 (d)(e)
10 100 1000 10000 F L U X ( m Jy ) FREQUENCY (MHz)UM 163 α = −0.82+/−0.03 (f)
Figure 9.
The spectral index α obtained from the fitted spectrum of the nuclear flux densities. Filled circles in the plots show the nuclear fluxes. Open squaresshow the NVSS flux densities. Filled triangle in the plot of UGC 6614 shows the jets / lobe flux density.c (cid:13) , 000–000 Mishra et. al.
NIR is detected from the centre of the galaxy and from a couple ofmassive star forming regions to the north-east of the centre of thegalaxy. On the other hand, UGC 2936 has hosted a core-collapsetype II supernova (SN 1991bd). We note that while the star form-ing disk of UGC 2936 is detected in the radio and NIR bands, onlyfaint NUV emission is detected from the centre of this galaxy (seeFigure 3 (a)). This could indicate an older star forming episode, saya few Gyrs ago which would be responsible for the NIR emissionand a more recent star forming episode which would be responsiblefor the radio emission and the detected supernova type II. Lack ofNUV emission could indicate that the later star formation episodemight have since been quenched. We note that it might be possibleto verify these results by a detailed study of the age of the stellarpopulations in these galaxies which is beyond the scope of this pa-per. Group membership is confirmed for UGC 1922 and UGC 4422whereas we could not find any information for UGC 2936 in litera-ture. While this is a qualitative picture, a more quantitative pictureof the multi-band emission might be able to result in a more com-plete picture of the timescales that the di ff erent diagnostics tracers.Another measure of the AGN versus starburst nature of the ra-dio emission is the FIR-radio correlation parameter q. The normalgalaxies have a value of q = We examined the environment of the sample galaxies. Four of thesample galaxies are reported in literature to have group member-ship; UGC 6614 (LDCE 829), UGC 4422 (LGG 159, LDCE 571),UM 163 (LDCE 1583), UGC 1922 (LDCE 163). LDCE (Low Den-sity Contrast Extended) groups have been catalogued by Crook etal. (2007, 2008) from 2MASS with the number of members rangingfrom 3 to 37.A study of small scale environment of GLSB galaxies(Bothun et al. 1993) has revealed that there is a deficit of galax-ies located within 0.5 Mpc and within a velocity of 500 kms − .The distance to the nearest neighbour for GLSB galaxies is about1.7 times farther than for HSBGs (Bothun et al. 1993). Contraryto this, Sprayberry et al. (1995) inferred from their study that thesmall scale environment of GLSB galaxies was similar to that ofHSBGs. Thus, it is not clear if the unevolved LSB disks are dueto the lack of nearby companions which would tidally trigger starformation.UGC 4422, which is catalogued as a member of LGG 159(Garcia 1993) is the only galaxy in our sample whose disk is brightin radio, NIR and NUV and hence is likely to be subjected to en-hanced and possibly continuous tidal interactions which play animportant role in the evolution of the galaxy. All the four galax-ies which are in groups show an extended UV disk in the GALEXdata indicating a recent burst of star formation. The tidal interactionin the group environs is likely responsible for this. Interestingly,UGC 1378 and UGC 2936, which are known not to have any groupmembership, the integrated radio emission appears to include largecontribution from a star forming disk. In UGC 2936 the observedfaint NUV emission is interpreted as indicating an older episode ofstar formation. The only isolated GLSB galaxy which shows brightNUV emission is Malin 2 and the source of the trigger of star for- mation in this galaxy is likely internal. We note that, the discussionhere supports the episodic star formation in GLSB galaxies withbursts of star formation followed by a quiescent phase as suggestedby Boissier et al. (2008) from their GALEX UV study of 13 LSBgalaxies. We have mapped the radio continuum emission at L band, 610,325 and 240 MHz of a sample of seven GLSB galaxies using theGMRT. All the galaxies host an optically identified AGN. Belowwe give the summary of our results and list our main conclusions. We detect compact radio emission from the centres of all thesample GLSB galaxies. The spectra of five galaxies (UGC 1922,UGC 2936, UGC 4422, Malin2 and UGC 6614) have spectral in-dices ranging from 0.12 to -0.44 and that of the galaxies UGC1378 and UM 163 exhibit a steeper spectrum. Two of the galaxiesUGC 6614 and UM 163 show extended emission associated withtheir nuclei but show no correlation with star formation traced byother diagnostics. We interpret the extended emission as being dueto the radio jets or lobes of the active nucleus. In UGC 6614 the ra-dio jet extends out to a radius of 6.8 kpc and in UM 163 it extendsout to 13 kpc. In both cases, the jet lies within the optical disks.Radio jets are relatively rare in spirals. Di ff use radio continuum emission associated with star for-mation in the disk is detected from the galaxies UGC 2936 andUGC 4422 at one of our observed frequencies. We used our highresolution maps at L band and the NVSS maps at 45 ′′ resolution toseparate the nuclear emission and disk emission. The radio emis-sion outside the nucleus was presumed as being due to star forma-tion and SFRs were found to range from 0.15 to 3.6 M ⊙ / yr. All the galaxies in our sample have been observed in theUV by GALEX and in the NIR by 2MASS thus allowing us to makea multi-wavelength study of this sample. All our sample galaxieshave bulges that are prominent in NIR. Extended UV disks are de-tected in five galaxies namely UGC 1922, UGC 4422, Malin 2,UGC 6614 and UM 163. NIR disk / bar emission is detected fromUGC 2936, UGC 4422 and UGC 1378. UGC 4422 is the only sam-ple galaxy which shows extended UV disks and NIR disk / bar emis-sion. We suggest that a recent burst of star formation has occurredin the five galaxies with extended UV disks out of which we findfour reside in group environments. We find that most of the radioemission in UGC 1922 and UM 163 arises in the active nucleus.Since these two galaxies show bright NUV emission, this suggestsa fresh star forming episode. The massive stars are yet to evolveinto supernovae and give rise to non-thermal emission. On the otherhand, hardly any NUV emission is detected from UGC 2936 butthe galaxy shows that more than 80% of its L band radio emissionarises from its star forming disk. This argues for a star formationepisode which is more than 10-100 million years old and has sincebeen quenched. It would be interesting to estimate stellar ages inthese galaxies to confirm this scenario. The study of these seven galaxies suggests that the environ-ment in which the galaxies are evolving is an important trigger forstar formation. It would be interesting to extend this study and ex-amine the environment of a large sample of GLSB galaxies whichis now possible with several catalogue of groups of galaxies and ofisolated galaxies. Finally, we speculate that the low surface bright-ness phase of evolution might be an important evolutionary step formost of the disk galaxies, especially for galaxies evolving in poorenvironment. c (cid:13) , 000–000 ow Surface Brightness Galaxies Table 6.
Star Formation Rate and FIR-Radio Correlation (q) for GalaxiesGalaxies SFR
NUV a
SFR
TIR L . GHzd
SFR . GHzd
SNR e logFIR qM ⊙ yr − M ⊙ yr − (WHz − ) M ⊙ yr − yr − − (Wm − )UGC 1378 .... .... 1.2 0.28 0.01 6.19 2.3UGC 1922 .... .... 3.1 0.75 0.03 .... ....UGC 2936 0.84 .... 10.0 2.45 0.12 28.42 2.2UGC 4422 .... 2.80 b c a SFR
NUV taken from Boissier et al. (2008) b SFR
TIR estimated from the total IR luminosity (L
TIR ) using the relation between L
TIR and SFR
TIR (Bell 2003) c SFR
TIR taken from Rahman et al. (2007) d L . GHz and SFR . GHz estimated from the di ff erence between NVSS flux density andGMRT nuclear flux density listed in Table 5 e Supernova rate and SFR . GHz estimated using the relation given in Condon (1992)
ACKNOWLEDGMENTS
We feel highly obliged to the referee for constructive commentswhich has improved the original manuscript. We thank the sta ff of the GMRT who made the observations possible. The GMRT isoperated by the National Centre for Radio Astrophysics (NCRA),Pune of the Tata Institute of Fundamental Research. AM wouldlike to thank the NCRA-TIFR, Pune and Indian Institute of As-trophysics, Bangalore for hospitality. This research has made useof NASA / IPAC Infrared Science Archive, the NASA / IPAC Extra-galactic Data base (NED), GALEX and 2MASS which is a NASAmission are operated by Jet Propulsion Laboratory, California In-stitute of Technology under contract with the National Aeronauticsand Space Administration. We also like to acknowledge the sitehttp: // skyview.gsfc.nasa.gov / . REFERENCES
Armstrong M., Schwartz M., 1999, IAU Circ., 7108, 1Auld R., de Blok W. J. G., Bell E., Davies J. I., 2006, MNRAS,366, 1475Beijersbergen M., de Blok W. J. G., van der Hulst J. M., 1999,A&A, 351, 903Bell E. F., 2003, ApJ, 586, 794Bell E. F., Barnaby D., Bower R. G., de Jong R. S., Harper D. A.,Hereld M., Loewenstein R. F., Rauscher B. J., 2000, MNRAS,312, 470Boissier S., Gil de Paz A., Boselli A., Buat V., Madore B., CheminL., Balkowski C., Amram P., Carignan C., van Driel W., 2008,ApJ, 681, 244Bothun G. D., Schombert J. M., Impey C. D., Schneider S. E.,1990, ApJ, 360, 427Bothun G. D., Schombert J. M., Impey C. D., Sprayberry D., Mc-Gaugh S. S., 1993, AJ, 106, 530Briggs D. S., 1995, BAAS, 27, 1444Burkholder V., Impey C., Sprayberry D., 2001, AJ, 122, 2318Comer´on S., Knapen J. H., Beckman J. E., Laurikainen E., SaloH., Mart´ınez-Valpuesta I., Buta R. J., 2010, MNRAS, 402, 2462Condon J. J., 1992, ARA&A, 30, 575Cornwell T. J., Perley R. A., 1992, A&A, 261, 353 Crook A. C., Huchra J. P., Martimbeau N., Masters K. L., JarrettT., Macri L. M., 2007, ApJ, 655, 790Crook A. C., Huchra J. P., Martimbeau N., Masters K. L., JarrettT., Macri L. M., 2008, ApJ, 685, 1320Das M., Boone F., Viallefond F., 2010, A&A, 523, A63Das M., O’Neil K., Vogel S. N., McGaugh S., 2006, ApJ, 651,853Das M., Reynolds C. S., Vogel S. N., McGaugh S. S., KanthariaN. G., 2009, ApJ, 693, 1300de Blok W. J. G., van der Hulst J. M., Bothun G. D., 1995, MN-RAS, 274, 235Galaz G., Dalcanton J. J., Infante L., Treister E., 2002, AJ, 124,1360Galaz G., Herrera-Camus R., Garcia-Lambas D., Padilla N., 2011,ApJ, 728, 74Gallimore J. F., Axon D. J., O’Dea C. P., Baum S. A., Pedlar A.,2006, AJ, 132, 546Garcia A. M., 1993, A&AS, 100, 47Geller M. J., Diaferio A., Kurtz M. J., Dell’Antonio I. P., FabricantD. G., 2012, AJ, 143, 102Gil de Paz A., Madore B. F., Boissier S., Swaters R., PopescuC. C., Tu ff s R. J., Sheth K., Kennicutt Jr. R. C., Bianchi L.,Thilker D., Martin D. C., 2005, ApJ, 627, L29Hinz J. L., Rieke M. J., Rieke G. H., Willmer C. N. A., MisseltK., Engelbracht C. W., Blaylock M., Pickering T. E., 2007, ApJ,663, 895Ho L. C., Filippenko A. V., Sargent W. L. W., 1997, ApJ, 487, 568Impey C., Bothun G., 1997, ARA&A, 35, 267Impey C. D., Sprayberry D., Irwin M. J., Bothun G. D., 1996,ApJS, 105, 209Kennicutt Jr. R. C., Keel W. C., Blaha C. A., 1989, AJ, 97, 1022Krause M., Fendt C., Neininger N., 2007, A&A, 467, 1037Laine S., Beck R., 2008, ApJ, 673, 128Mapelli M., Moore B., Ripamonti E., Mayer L., Colpi M., Gior-dano L., 2008, MNRAS, 383, 1223Martin D. C., Fanson J., Schiminovich D., Morrissey P., FriedmanP. G., Barlow T. A., Conrow T., 2005, ApJ, 619, L1Mayer L., Wadsley J., 2004, MNRAS, 347, 277McGaugh S. S., 1994, ApJ, 426, 135McGaugh S. S., de Blok W. J. G., 1998, ApJ, 499, 41 c (cid:13) , 000–000 Mishra et. al.
McGaugh S. S., Schombert J. M., Bothun G. D., 1995, AJ, 109,2019Mei L., Yuan W.-M., Dong X.-B., 2009, Research in Astronomyand Astrophysics, 9, 269Mihos J. C., McGaugh S. S., de Blok W. J. G., 1997, ApJ, 477,L79Morelli L., Corsini E. M., Pizzella A., Dalla Bont`a E., Coccato L.,M´endez-Abreu J., Cesetti M., 2012, MNRAS, 423, 962Mueller J., Gunn J., Oke J., 1989, IAU Circ., 4888, 1Mueller J., Mendenhall J. D., Mateo M., Reid I., 1991, IAU Circ.,5370, 3Naik S., Das M., Jain C., Paul B., 2010, MNRAS, 404, 2056O’Neil K., Oey M. S., Bothun G., 2007, AJ, 134, 547O’Neil K., Schinnerer E., 2003, ApJ, 588, L81Pickering T. E., Impey C. D., van Gorkom J. H., Bothun G. D.,1997, AJ, 114, 1858Pustilnik S. A., Martin J.-M., Tepliakova A. L., Kniazev A. Y.,2011, MNRAS, 417, 1335Raha N., Sellwood J. A., James R. A., Kahn F. D., 1991, Nature,352, 411Rahman N., Howell J. H., Helou G., Mazzarella J. M., BuckalewB., 2007, ApJ, 663, 908Ramya S., Prabhu T. P., Das M., 2011, MNRAS, 418, 789Rosenbaum S. D., Krusch E., Bomans D. J., Dettmar R.-J., 2009,A&A, 504, 807Schombert J., 1998, AJ, 116, 1650Schombert J., McGaugh S., Maciel T., 2013, AJ, 146, 41Skrutskie M. F., Cutri R. M., Stiening R., Weinberg M. D., Schnei-der S., 2006, AJ, 131, 1163Sprayberry D., Impey C. D., Bothun G. D., Irwin M. J., 1995, AJ,109, 558Swarup G., Ananthakrishnan S., Kapahi V. K., Rao A. P., Subrah-manya C. R., Kulkarni V. K., 1991, CURRENT SCIENCE V.60,NO.2 / JAN25, P. 95, 1991, 60, 95Thilker D. A., Bianchi L., Meurer G., Gil de Paz A., Boissier S.,Madore B. F., 2007, ApJS, 173, 538van den Hoek L. B., de Blok W. J. G., van der Hulst J. M., de JongT., 2000, A&A, 357, 397van der Hulst J. M., Skillman E. D., Smith T. R., Bothun G. D.,McGaugh S. S., de Blok W. J. G., 1993, AJ, 106, 548Yun M. S., Reddy N. A., Condon J. J., 2001, ApJ, 554, 803 c (cid:13)000