Halpha imaging survey of Wolf-Rayet galaxies: morphologies and star formation rates
aa r X i v : . [ a s t r o - ph . GA ] J un Mon. Not. R. Astron. Soc. , 1– ?? (2014) Printed 26 October 2018 (MN L A TEX style file v2.2) H α imaging survey of Wolf-Rayet galaxies:morphologies and star formation rates S. Jaiswal , ⋆ and A. Omar Aryabhatta Research Institute of Observational Sciences (ARIES), Manora Peak, Nainital, 263002, India Pt. Ravishankar Shukla University, Raipur, 492010, India
Accepted —. Received —; in original form —
ABSTRACT
The H α and optical broadband images of 25 nearby Wolf-Rayet (WR)galaxies are presented. The WR galaxies are known to have the presence ofa recent ( ≤
10 Myr) and massive star formation episode. The photometricH α fluxes are estimated, and corrected for extinction and line contamina-tion in the filter pass-bands. The star formation rates (SFRs) are estimatedusing H α images and from the archival data in the far-ultraviolet (FUV),far-infrared (FIR) and 1.4 GHz radio continuum wave-bands. A comparisonof SFRs estimated from different wavebands is made after including similardata available in literature for other WR galaxies. The H α based SFRs arefound to be tightly correlated with SFRs estimated from the FUV data. Thecorrelations also exist with SFRs estimates based on the radio and FIR data.The WR galaxies also follow the radio-FIR correlation known for normal starforming galaxies, although it is seen here that majority of dwarf WR galaxieshave radio deficiency. An analysis using ratio of non-thermal to thermal radiocontinuum and ratio of FUV to H α SFR indicates that WR galaxies havelesser non-thermal radio emission compared to normal galaxies, most likelydue to lack of supernova from the very young star formation episode in theWR galaxies. The morphologies of 16 galaxies in our sample are highly sug-gestive of an ongoing tidal interaction or a past merger in these galaxies. Thissurvey strengthens the conclusions obtained from previous similar studies in-dicating the importance of tidal interactions in triggering star-formation inWR galaxies. c (cid:13) Jaiswal & Omar
Key words: galaxies: interaction – galaxies: starburst – galaxies: Wolf-Rayet– galaxies: star formation rates – galaxies: H α photometry Wolf-Rayet (WR) galaxies are a subset of emission-line and H ii galaxies whose integratedoptical spectra exhibit broad emission line features (He ii λ iv λ - 10 ) ofWR stars in these galaxies (e.g., Kunth & Sargent 1981; Kunth & Schild 1986). The mostmassive O-type stars (M ≥
25 M ⊙ for solar metallicity) become WR stars after 2 to 5 Myrfrom their birth, spending only a short time (t W R ≤ <
10 Myr) ongoing star-burst in a galaxy (e.g., Schaerer et al. 1999). The WR galaxies,therefore, offer an unique opportunity to study the onset of star formation as well as theconditions, which may be responsible for triggering star-formation in galaxies (Schaerer &Vacca 1998). WR galaxies are quite rare in the nearby Universe as the WR phase is a veryshort-lived stage. The first identification of the WR features in a galaxy was in the bluecompact dwarf galaxy He 2-10 (Allen et al. 1976). Conti (1991) compiled a WR galaxycatalogue having 37 objects. Later, Schaerer et al. (1999) made a catalogue of 139 WRgalaxies. The most recent catalogue of WR galaxies is compiled by Brinchmann et al. (2008)using the
Sloan Digital Sky Survey ( SDSS ) data. It lists 570 WR galaxies and a further1115 suspected WR galaxies. Gil de Paz et al. (2003) presented B , R and H α images of atotal of 114 nearby blue compact dwarf (BCD) galaxies, many of those were identified asWR galaxies with high star formation rates.The star-formation is considered a fundamental parameter for evolution of galaxies. Thestar-formation in a galaxy is measured in terms of star formation rate (SFR). The SFRis estimated from the number of Lyman continuum photons ( N LyC ), which are emittedmainly by the young massive stars. The N LyC can be measured from the continuum fluxin far ultraviolet (FUV) or blue wave-bands (e.g., Gallagher et al. 1984; Bell & Kennicutt2001; Salim et al. 2007; Murphy et al. 2011), far-infrared (FIR, e.g., Kennicutt et al. 1987; ⋆ E-mail: [email protected] c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies α λ ii ] λ α emission directly provides an estimate of the ionizing flux fromthe young massive stars, and is amongst the strongest optical emission lines in star forminggalaxies. The H α emission flux is less sensitive to dust extinction and metallicity comparedto the [O ii ] line. Other indicators such as X-ray continuum (e.g., Ranalli et al. 2003; Mineoet al. 2011) and mid-infrared PAH (polycyclic aromatic hydrocarbon) emission (e.g., Roussel2001; Calzetti et al. 2007; Calzetti 2011) can also be used to get an estimate for SFR ingalaxies.These SFR indicators often predict different values of SFR due to various uncertaintiesarising from extinction correction, stellar evolution models, magnetic field strengths, diffu-sion and re-processing time scales of stellar continuum photons in galaxies (Calzetti 2013).The FUV continuum emission from the young and massive stars can be sensitive to starformation events over the time scales of a few tens of Myr. The diffuse synchrotron radioemission tracing its origin in loss of energy from relativistic cosmic electrons acceleratedin supernovae events and diffused out to kilo-parsec scales is sensitive to star formationevents typically during the last 100 Myr. The H α emission line, on the other hand, providesinformation on the most recent ( ∼
10 Myr) star formation in galaxies (Kennicutt 1998a;Murphy et al. 2011). In case of star formation taking place in highly dust-obscured regions,accuracy of extinction correction becomes very important in getting SFR estimates from theFUV and optical bands. Although, the radio emission is not affected by dust, uncertaintieson strength of magnetic field and contamination due to active galactic nuclei (AGN) maymake radio flux a relatively less reliable estimator for the SFR in galaxies. Investigationson correlations between different SFR tracers are important to understand various physicalprocesses in the interstellar medium (ISM). Such correlations also help in constraining SFRsin galaxies. Due to very recent star formation episode in WR galaxies, the star formationage may have subtle effects on the observable fluxes at different wavebands, and hence onSFR correlations. Such effects have not been studied very well over a large range of SFRvalues in WR galaxies. In a study of 20 WR galaxies, L´opez-S´anchez (2010) found that SFRderived from the H α photometry of WR galaxies yield statistically significant correlationswith other tracers such as FUV, FIR, X-ray and radio continuum. c (cid:13) , 1–, 1–
10 Myr) star formation in galaxies (Kennicutt 1998a;Murphy et al. 2011). In case of star formation taking place in highly dust-obscured regions,accuracy of extinction correction becomes very important in getting SFR estimates from theFUV and optical bands. Although, the radio emission is not affected by dust, uncertaintieson strength of magnetic field and contamination due to active galactic nuclei (AGN) maymake radio flux a relatively less reliable estimator for the SFR in galaxies. Investigationson correlations between different SFR tracers are important to understand various physicalprocesses in the interstellar medium (ISM). Such correlations also help in constraining SFRsin galaxies. Due to very recent star formation episode in WR galaxies, the star formationage may have subtle effects on the observable fluxes at different wavebands, and hence onSFR correlations. Such effects have not been studied very well over a large range of SFRvalues in WR galaxies. In a study of 20 WR galaxies, L´opez-S´anchez (2010) found that SFRderived from the H α photometry of WR galaxies yield statistically significant correlationswith other tracers such as FUV, FIR, X-ray and radio continuum. c (cid:13) , 1–, 1– ?? Jaiswal & Omar
The star formation trigger in WR galaxies is also important to understand. Several stud-ies on normal star-forming galaxies indicate that gravitational tidal interactions and mergersof galaxies play a major role in galaxy evolution and triggering star formation in galaxies(Larson & Tinsely 1978; Koribalski 1996; Kennicutt 1998b; Nikolic et al. 2004; Joseph &Wright 1985; Solomon & Sage 1988; Sanders & Mirabel 1996; Genzel et al. 1998; Omar &Dwarakanath 2005). The hierarchical growth models of galaxies support the formation oflarger structures like giant spiral and massive elliptical galaxies through mergers and ac-cretion of smaller structures like dwarf galaxies (e.g., Shlosman 2013; Amorisco et al. 2014;Deason et al. 2014), implying that tidal interactions should be fairly common. A causal linkbetween star formation and mergers or interactions has been seen in several blue compactdwarf galaxies (e.g., Bravo-Alfaro et al. 2004; Bekki 2008; L´opez-S´anchez et al. 2012; Ashley2014). The tidal interactions between galaxies can be traced in several ways. The morpho-logical features such as tidal tails and plumes detected in the H i i α line imaging. A misalignment between the H α disk and thestellar disk can also be an indicator of interaction as seen in the WR dwarf galaxy MRK 996(Jaiswal & Omar 2013). The interaction features in galaxies can also be traced by studyingkinematics of the ionized gas or neutral H i gas (e.g., L´opez-S´anchez et al. 2004a,2004b,2006;L´opez-S´anchez & Esteban 2009; L´opez-S´anchez 2010).Despite the importance of studying interactions in WR galaxies, very few detailed stud-ies of WR galaxies have been carried out. M´endez & Esteban (2000) performed deep opticalimaging and spectroscopy on a sample of WR galaxies, and found that the star-burststriggered in low mass WR galaxies could be due to interactions with dwarf galaxies. L´opez-S´anchez & Esteban (2008, 2009, 2010a, 2010b) and L´opez-S´anchez (2010) found that ma-jority of galaxies in a sample of 20 WR galaxies were clearly interacting or merging withlow luminosity dwarfs objects or intergalactic H i clouds. It is suggested by these authorsthat interacting or merging nature of WR galaxies can be detected only when both deep, c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies α and r -band images of 25 WR galaxies are presented here. A total of 23 out of25 WR galaxies in the present sample were selected from SDSS, which provides broad-bandphotometry and spectroscopic data in the visible range of the electromagnetic spectrum(see, Abazajian et al. 2009 and the references therein). Since the SDSS spectrum has beenobtained using a fiber slit of 3 ′′ diameter and galaxy sizes of WR galaxies are several arc-min in our sample, total SFR can not be reliably derived from the SDSS spectroscopicdata alone. The narrow-band H α photometry maps full extent of galaxies, and therefore,provides total SFR. The H α flux calculated from the photometric observations needs tobe corrected for contaminations by other emission lines in the filter passbands apart fromthe galactic and internal extinction corrections. These corrections need a knowledge of thephysical conditions in the ISM, and hence require spectroscopic data. The corrections tothe H α flux were estimated in this paper primarily using the SDSS spectroscopic data. Wehave combined our sample of 25 WR galaxies with the database on 20 WR galaxies fromL´opez-S´anchez (2010) making the total sample size up to 45 galaxies. The H α -derived SFRestimates are compared with those estimated from the FUV, FIR and radio continuumluminosities from the archival data. The radio-FIR correlation is also constructed usingthe archival data from IRAS (Infrared Astronomical Satellite) at FIR wavelengths, and the
FIRST (Faint Images of the Radio Sky at Twenty-cm) and the
NVSS (NRAO VLA SkySurvey) images at 1.4 GHz taken with the Very Large Array (VLA). The H α and SDSSimages were used to investigate interaction features in WR galaxies. c (cid:13) , 1–, 1–
NVSS (NRAO VLA SkySurvey) images at 1.4 GHz taken with the Very Large Array (VLA). The H α and SDSSimages were used to investigate interaction features in WR galaxies. c (cid:13) , 1–, 1– ?? Jaiswal & Omar
The WR features are found in nearly all morphological types of galaxies, ranging from low-mass dwarf galaxies and irregular galaxies to massive spirals, luminous mergers, infraredluminous galaxies and Seyfert galaxies (Ho et al. 1995; Heckman & Leitherer 1997; Zhanget al. 2007). We used the catalogs of WR galaxies prepared by Schaerer et al. (1999) andBrinchmann et al. (2008) to construct a sample. We selected galaxies up to ∼
25 Mpc distance(with H = 75 km s − Mpc − ), so that galaxies can be observed with the H α filters availableon the telescopes used. We further restricted our sample with the declination > − ◦ , sothat galaxies can be observed for sufficient integration time from the telescopes. We finallyselected a total of 25 galaxies to carry out this study. The selected galaxies are brighter than17 mag in the SDSS r -band, except NGC 2799 and UM 3111 whose SDSS r magnitudes are18.6 and 17.8 respectively.The basic properties of the galaxies in our sample are given in Table 1, which is con-structed using NASA/IPAC extragalactic database (NED). In this table, the SDSS r -bandmagnitudes for galaxies UGCA 116, UGCA 130, UM 439 and I SZ 59 are extrapolated fromthe Johnson B and Cousins R band magnitudes (Gil de Paz et al. 2003) with the help ofLupton transformation relations published on the SDSS DR4 website, as these galaxies werenot observed in the SDSS. The extinction and metallicity parameters for the selected galax-ies are given in Table 2. For most of the galaxies, the mixed oxygen abundance estimates asgiven in Brinchmann et al. (2008) are provided here. Figure 1 shows the histograms of linearsizes and oxygen abundances of the selected galaxies. The linear size distribution indicatesthat the sizes of galaxies in the sample are <
25 kpc, and majority of them are <
10 kpc. Weterm galaxies with size <
10 kpc as small and other galaxies as large. The oxygen abundancehistogram shows that the selected WR galaxies have metallicities between 8 and 9. A verticalline showing the solar metallicity, 12 + log(O/H) = 8.69 (Asplund et al. 2009) is drawn inthis figure. The majority of galaxies in our sample have sub-solar metallicities. The colorcomposite images made using the SDSS g , r and i bands images of the galaxies are shownin Figure 2. The blue regions here show star forming regions. The H α and r -band observations of 21 galaxies were carried out using the 1.3-meter Dev-asthal Fast Optical Telescope (DFOT; longitude = 79 o ′ ′′ E, latitude = 29 o ′ ′′ N, c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies ∼ . ′′ pixel − . The data were recorded on a thermo-electrically cooled ( − ◦ C) CCD (chargecoupled device) camera having a back-illuminated E2V chip of 2048 × µ m. The CCD covers a circular field of view of diameter ∼ ′ on the sky. The CCD readoutnoise is nearly 7 e − at 1 MHz speed with a gain of 2 e − ADU − . A brief overview of thistelescope is provided in Sagar et al. (2011) and Jaiswal & Omar (2013). The observations ofthe remaining 4 galaxies were carried out using the 2-meter telescope of the Inter UniversityCentre for Astronomy and Astrophysics (IUCAA) Girawali Observatory (IGO; longitude= 73 o ′ E, latitude = 19 o ′ N, altitude ∼ × µ m pixel size. The CCD used for imaging providesa field of view of 10 . ′ × . ′ on the sky corresponding to a plate scale of 0 . ′′ pixel − . Thegain and readout noise of the CCD camera are 1.5 e − ADU − and 4 e − , respectively.The observations were carried out mostly in the dark nights near the new moon pe-riods. The observation log is provided in Table 3. The observations were carried out inthe H α narrow-band filter and the SDSS- r or Cuisine R filter available on the telescopes.The central wavelength and FWHM (Full Width Half Maxima) of the H α filters used, andH α sensitivities in the observations are given in Table 4. Here, the H α sensitivity was calcu-lated for 3 σ detection, and λ H α is the observed wavelength for the red-shifted H α emissionand λ is the central wavelength of the H α filter. The pass-band transmittance curves forthe H α filters used are shown in Figure 3. The total integration time was divided into ∼ α bandwas typically between 100 and 200 minutes while that in the r band was typically between10 and 30 minutes. At least one standard spectrophotometric star selected from Oke (1990)was observed a few times at different airmass values every night. The standard star was usedto obtain the photometric calibration and the atmospheric extinction parameters. We alsocarried out spectroscopic observations of two galaxies, namely KUG 1013+381 and I SZ 59,in our sample using the 2-meter Himalayan Chandra Telescope (HCT), Hanle which is op-erated by the Indian Institute of Astrophysics (IIA), Bangalore, India. The details of thespectroscopy observations will be presented elsewhere. Here, the results from these spec- c (cid:13) , 1– ?? Jaiswal & Omar troscopy observations are used to apply the internal reddening corrections to the H α fluxfor these two galaxies. The CCD images were cleaned following the standard procedures of bias subtraction andflat-fielding using the CCDPROC task of IRAF (Image Reduction and Analysis facility) soft-ware developed by
National Optical Astronomy Observatory (NOAO) . The dark currents inthe CCDs used are negligible and hence no dark subtraction was applied. The cosmic rayswere removed interactively using the IRAF tasks COSMICRAYS and CREDIT within theCRUTIL package. The observed frames for a galaxy were aligned using the IRAF tasks GE-OMAP and GEOTRAN. As the image frames were taken at different values of airmass, theimages were corrected for the atmospheric extinction before combining the aligned frames.The noise weighted mean of these frames was estimated to get the combined frame for aparticular filter. The weighting factor multiplied to each frame was 1 /σ , where σ is thebackground sky rms for the frame. The H α emission-line flux measured by the narrow-band filter includes both the line fluxand the continuum flux. This continuum emission is dominated by stars in the galaxy. Itis therefore necessary to subtract the continuum in order to estimate the H α emission-lineflux. The underlying continuum in the H α filter was subtracted using the standard proceduregiven by Waller (1990) and Spector et al. (2012). In this method, the underlying continuumemission is estimated from the nearest wide-band such as r -band filter, which is significantlywider compared to the narrow band. The main assumption here is that the galaxy continuumflux per unit wavelength does not differ much across the narrow-band and the wide-bandfilters. The continuum emission in the wide-band filter is scaled to that in the narrow-bandfilter by using a scaling factor, called the Wide to Narrow Continuum Ratio (WNCR),defined as the ratio of the continuum count rate in the wide-band filter to the continuumcount rate in the narrow-band filter. The count rates are determined using unsaturated,non-variable, isolated field stars in the wide and narrow band images. Here, it is importantthat the field stars should not have any strong emission or absorption line within the wide- c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies α and r -band filtersis given in Table 4. The WNCR value is used to scale down the r -band image so that thecounts per second of field stars in the r -band and the H α -band are almost identical. Thecontinuum subtracted H α images were obtained by subtracting the scaled r -band imagesfrom the H α -band images. After continuum subtraction, all field stars should normally becompletely removed from the image. However, in practice, some residual emission may beseen at the locations of bright stars due to imperfections in the assumptions made and somevariations in the seeing conditions during the observations. Our observations in the H α and r -bands were always taken during the same night, which minimized effects due to extremevariations in inter-night seeing and in most of the cases, the continuum subtraction processworked very well. Only in a few frames, some residual continuum emission was visible atlocations of very bright stars in the frame. The typical error in the estimates of WNCRvalues is 5%. The images were registered for the equatorial coordinates in the J2000 epoch using the
TwoMicron All Sky Survey (2MASS) or SDSS r -band images of the same region. In this process,the World Coordinate System (WCS) solutions are calculated using the pixel coordinates ofa set of stars in the target frame and the WCS coordinates of the same stars in the referenceframe using the IRAF task CCMAP. On applying the WCS solutions on the target frames,the astrometric uncertainty is found to be less than 0 . ′′ . The same astrometric calibrationwas applied to the continuum subtracted H α image. These H α and r -band images were madein publishable format using IGI (Interactive Graphics Interpreter) package of STSDAS inIRAF. α flux determination and calibration In order to obtain integrated H α flux, the polygon aperture photometry was performedon the continuum subtracted H α images using the IRAF task POLYPHOT. The size ofthe aperture was selected such that it included all the star-forming regions as well as diffuseH α emission from the galaxy. We have not attempted here to measure H α flux for individualstar forming regions and diffuse H α emission separately. The average sky background valueper pixel was measured close to the galaxy. The H α flux was calibrated using observations c (cid:13) , 1– ?? Jaiswal & Omar of spectrophotometric standard stars taken at different values of airmass. The atmosphericextinction was estimated from the observations of these standard stars. The flux (counts s − )of the standard star was estimated using the DAOPHOT package in IRAF. The expectedH α flux (erg s − cm − ) of the standard star was calculated by integrating the product ofthe calibrated spectrum of standard star and the filter transmission curve with respect tothe wavelength. The instrumental response was obtained by dividing the expected H α flux(erg s − cm − ) of the standard star with its observed H α flux (counts s − ) at zero airmass.The instrumental responses for the combinations of telescopes and filters are given in Table 4.The calibrated H α flux for a galaxy was estimated by multiplying its observed H α flux withthe instrumental response. The H α flux needs to be further corrected for an attenuation inthe H α filter transmission due to the shift in the emitted wavelength of the H α emission lineat the redshift of the galaxy. The narrow-band filters normally have maximum transmissionat the rest frame wavelength of the H α emission. The narrow-band filter response curve isalso dependent on the focal ratio of the telescope and requires a correction. The effect of thetelescope focal ratio is important only for the 1.3-meter DFOT, since the filters are placedhere in the fast (f/4) converging beam. Both of these corrections were estimated in Jaiswal& Omar (2013) and were applied here.The calibrated H α flux was corrected for line contaminations in the filter passbands,the Galactic extinction and the internal extinction. The narrowband (H α ) filters containsignificant flux from [N ii ] λλ r ) filter, in addition, will con-tain significant flux from [O i ] λ ii ] λ ii ] λ ii ] λ ii ] λ α λ α line is a con-taminating line in the r band for estimating continuum flux. The H α line flux should becorrected for these line contaminations. The fluxes of the emission lines contaminating thefilter passbands were estimated from the SDSS spectroscopic data. It is assumed here thatthe line ratios are constant everywhere in the galaxy. The galactic (foreground) extinctioncorrection was determined using E f ( B − V ) values based on Schlafly & Finkbeiner (2011)recalibration of the Schlegel, Finkbeiner & Davis (1998) extinction map. These values weretaken directly from NED. Assuming an intrinsic Case-B recombination ratio of 2.86 (Oster-brock 1989) for H α /H β ratio, valid for an ionized gas with an electron temperature of 10 Kand an electron density of 100 cm − ), and the Cardelli, Clayton & Mathis (1989) extinctioncurve with R V = 3 .
1, the internal reddening of the ionized gas can be expressed in terms ofthe observed H α /H β flux ratio (corrected for the Galactic extinction) as: c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies E g ( B − V ) = 2 . × log (cid:18) f Hα /f Hβ . (cid:19) (1)The values of E f ( B − V ) and E g ( B − V ) for each galaxy are listed in Table 2. The E ( B − V ) values can be changed into the extinction coefficients at the H α wavelengthusing the Cardelli, Clayton & Mathis (1989) extinction curve with R V = 3 . A Hα =2 . E ( B − V ). The H α and H β fluxes for several galaxies are taken from the MPA-JHU(Max-Planck-Institute for Astrophysics and Johns Hopkins University) emission line analysisdatabase for the SDSS data release-7 (DR7), which were already corrected for the foregroundextinction using O’Donnell (1994) extinction law. Therefore, the E ( B − V ) value calculatedusing these line fluxes will only provide the internal extinction without a contribution fromforeground extinction due to the Milky-way. Therefore, we applied the foreground reddeningcorrections separately for those galaxies, for which MPA-JHU databases were used. Theratio for the galaxies MRK 475, UGCA 116 and UGCA 130 were taken from the literature.We estimated the H α /H β flux ratio for galaxies I SZ 59 and KUG 1013+381 from ourspectroscopic observations. The H α /H β flux ratios for these galaxies (MRK 475, UGCA 116,UGCA 130, I SZ 59 and KUG 1013+381) provide sum of foreground and internal extinctions. α flux estimates The accuracy of H α flux estimates from the narrow-band H α imaging depends on the ac-curate removal of the underlying continuum. The error in WNCR gives main contributionto the total error (Spector et al. 2012). We used 10 −
15 foreground stars to calculate theWNCR value. The residual flux at the locations of bright field stars after subtraction of thescaled continuum is similar to error in WNCR (below 5%), the errors in continuum subtrac-tion for galaxies can be fixed at this value. In some cases (UM 311, NGC 941, NGC 1087and UGCA 130), flat-fielding was not perfect and a large-scale residual gradient was seen inthe cleaned H α images. However, we always estimated local sky background around galax-ies, in order to minimize effects of improper flat-fielding on flux estimates. These effectsare expected to be insignificant compared to other errors. In addition to the WNCR error,there will be some additional sources of observational errors arising from uncertainties in theestimates of instrumental response, narrow-band filter transmission curve and atmosphericextinction. It is difficult to make an exact estimate for errors in the H α flux in individualcases. We believe that typical errors in H α flux corrections remain below 10%. The finalerrors in the H α flux estimates are typically between 10% and 20%. c (cid:13) , 1– ?? Jaiswal & Omar
The star formation rates in galaxies can be well constrained by multi-wavelength data. In thepresent study, we have used the archival data at FUV, FIR and 1.4 GHz radio bands to makeindependent estimates for SFR. The FUV data is taken from the
Galaxy Evolution Explorer (GALEX) survey. The details of the GALEX telescope, detectors and data products are givenin Morrissey et al. (2007). The GALEX telescope has two photometric bands, FUV (1344-1786 ˚A; λ eff = 1528 ˚A) and NUV (1771-2831 ˚A; λ eff = 2310 ˚A). We estimated FUV fluxdensities for galaxies in our sample from the GALEX FUV images using POLYPHOT taskof IRAF. The measured counts per second ( CP S ) in the FUV image is converted into fluxdensity using the calibration relation: f F UV (erg s − cm − ˚A − ) = 1 . × − × CP S . Thisflux density is corrected for the Galactic extinction using E f ( B − V ) values in Table 2 andfor the internal extinction of the stars using E s ( B − V ) values. Following Calzetti (1997), thecolor excess of the stellar continuum E s ( B − V ) is related to the color excess of the ionizedgas E g ( B − V ) as: E s ( B − V ) = (0 . ± . E g ( B − V ). The total E ( B − V ) will give theextinction A F UV = 8 . E ( B − V ) using the Cardelli, Clayton & Mathis (1989) extinctioncurve with R V = 3 .
1. The corrected FUV flux densities for the sample galaxies are given inTable 5.The FIR data is taken from the IRAS survey (Neugebauer et al. 1984). The IRASmission performed a low-resolution all sky survey at 12 µ m, 25 µ m, 60 µ m and 100 µ m.We have used the 60 µ m and the 100 µ m FIR flux densities as given by NED to get anestimate of FIR emission from the sample galaxies. These flux densities are given in Table 5.The 1.4 GHz radio continuum data is taken from the FIRST and NVSS databases. TheFIRST survey (Becker, White & Helfand 1995) images have a typical rms of 0.15 mJybeam − and an angular resolution of 5 ′′ . On the other hand, NVSS (Condon et al. 1998)images have a typical rms of 0.45 mJy beam − and a resolution of 45 ′′ . The NVSS imagesare more sensitive to extended emission. We used the FIRST data for small galaxies andthe NVSS data for large galaxies. The 1.4 GHz flux densities from the NVSS are takendirectly from NED. We convolved the FIRST images with 20 ′′ × ′′ beam size to makeimage resolution comparable to the source size, before estimating flux densities. The AIPS(Astronomical Image Processing System; Greisen 2003) developed by the National RadioAstronomy Observatory (NRAO) was used for analysis of radio images. c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies The grey-scale H α and r -band images of all the galaxies are presented in Figure 4. A lin-ear scale in kilo-parsec is shown at the bottom of the each image. The H α sensitivitiesof the images presented here are typically ∼ − erg s − cm − arcsec − . The observedsample was conveniently divided into two major classes: large spiral galaxies and dwarf orsmall galaxies. The large galaxies include NGC 1087, NGC 2799, NGC 3294, NGC 3381,NGC 3423, NGC 3430, NGC 3949, NGC 4389 and NGC 0941. On the other hand, thedwarf or small galaxies include CGCG 041-023, IC 0225, IC 2524, IC 2828, IC 3521,I SZ 059, KUG 1013+381, MRK 0022, MRK 0475, SBS 1222+614, UGC 09273, UGCA 116,UGCA 130, UM 311 and UM 439. The H α flux estimates for all the galaxies of our sampleare provided in Table 6. The second column ( F Hα ) of this table lists the calibrated but un-corrected H α flux. The subsequent columns show corrected H α flux in sequential steps as:(i) F iHα for line (e.g., N ii ii α filter,(ii) F iiHα for line (e.g., [O i ] 6300, [N ii ] 6548, H α ii ] 6584, [S ii ] 6717 and [S ii ] 6731)contamination in the wide-band r filter, and finally, (iii) F iiiHα was obtained by correcting F iiHα for the galactic extinction and the internal galaxy extinction using E ( B − V ) values. The star formation rates in galaxies can be estimated using fluxes at different wavelengths,viz., FUV, optical, IR and radio (e.g., Kennicutt 1998a; Buat et al. 2002; Rosa-Gonz´alez etal. 2002; Hopkins et al. 2003; Murphy 2011). The reliability of each of these indicators hasbeen widely discussed (e.g., Condon 1992; Kennicutt 1998a; Calzetti et al. 2007; Salim et al.2007). The complexity of the astrophysics underlying each method leads to a considerabledegree of uncertainty in its use as an estimator of the star formation rate. Therefore, it isuseful to estimate SFR using all possible methods. The SFR is calculated by multiplyingluminosity in a wave-band by a calibrated numerical factor. The numerical factor for eachband is computed using stellar evolutionary synthesis models and certain assumptions onthe physical processes in the ISM. There are slight differences in the conversion factors fora wave-band investigated in different research works. These minor differences arise due tousage of different stellar evolution models, atmosphere models, metal abundances, initialmass functions (IMF) and star formation histories. Such variations have been discussedpreviously (e.g., Kennicutt 1998a; Murphy et al. 2011). For instance, the conversion factor c (cid:13) , 1– ?? Jaiswal & Omar between SFR and H α luminosity given by Calzetti et al. (2007) is 0.67 times of that given byKennicutt (1998b). The calibration constant is also dependent on the age of the most recentstar formation event and star formation history in galaxies. Over a time scale of 100 Myr,the differential changes in the calibration constants are small. However, such effects may bevery important for young star-bursts like WR galaxies. A detailed description of conversionfactor and its sensitivity to different parameters is beyond the scope of this paper. A briefoverview of methods used to convert luminosities to SFR is provided in the Appendix. Theluminosities at different wave-bands and the corresponding estimates of SFRs are presentedin Table 7. This table also lists an assumed SFR estimated as a median value, obtained fromall the estimates of SFR. α and r -band morphologies The following notes summarize morphological features (see Figure 2 and Figure 4) seen foreach galaxy in our sample:
UM 311
UM 311 (shown by label ‘A’ in Figure 4) is considered as a giant H ii region complex inthe spiral galaxy NGC 450 or a dwarf H ii galaxy interacting with NGC 450. The opticalobservations has also been presented in Karthick et al. (2014). It has not been possible tofavour any one scenario over the other based on optical observations. Another smaller spiralgalaxy UGC 807 in the field is a background galaxy, and does not form a physical pair withNGC 450. We looked at high resolution 1.4 GHz image from the FIRST survey and noticedthat the radio emission is resolved primarily into four knots. The radio emission is not de-tected from the position of UM 311, but detected from two other nearby knots, which are alsodetected in H α . The FIR emitting regions seen in the Herschel images are coincident withthe radio emitting knots. We also notice that this H ii region complex is in a ring shape. InSDSS images, this complex is the bluest among other H ii regions in the galaxy. It leaves uswith an interesting possibility that UM 311 may be a dwarf galaxy interacting with NGC 450. IC 225
IC 225 is classified as a dwarf elliptical (dE) galaxy. The broad-band SDSS optical imagesrevealed two nuclei separated by 1.4 ′′ - a central core along with a blue region, which disap- c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies α emission is seen centrally concentratedin our images, and the nucleus is not resolved. No radio emission is detected from this galaxy. NGC 941
NGC 941 is a barred spiral galaxy with multiple spiral arms. It forms a non-interacting pairwith NGC 936, at 12.6 ′ separation (de Vaucouleurs et al. 1976). The star forming H α emit-ting regions are seen spread over the disk with a lopsided central concentration of brightH ii regions with a bar like feature. NGC 1087
NGC 1087 is a face-on large spiral galaxy with multiple spiral arms and a small bar. It is amember of NGC 1068 Group (de Vaucouleurs et al. 1976) and has a close companion galaxyNGC 1090. Blackman (1980) considered NGC 1087 as an isolated galaxy due to the largevelocity difference between NGC 1087 ( v = 1517 km s − ) and NGC 1090 ( v = 2760 km s − ).Martin & Friedli (1997) quoted two asymmetries in this galaxy, namely a large (13 ◦ ) mis-alignment between the H α bar and the stellar bar, and more star formation regions in thenorth, compared to the south. Hummel et al. (1987) found extended 1.4 GHz radio emissionwith nearly 1 ′ . α emission is seen from the nuclear bar.Several bright H ii regions are also seen along the spiral arms. This galaxy is highly lopsided. UGCA 116
The BCD galaxy UGCA 116 is a cometary type galaxy having a ‘head-tail’ structure (e.g.,Cair´os et al. 2001). Such an unusual morphology is expected from a merger between twosmall galaxies (Baldwin, Spinrad & Terlevich 1982; Brinks & Klein 1988). Optical and radioobservations (e.g., Sage et al. 1992; Deeg et al. 1997; van Zee et al. 1998) shows an extraor-dinary star formation, which indicates that UGCA 116 might be observed at the peak of itsstar forming episode caused by a rare merger between two gas-rich dwarf galaxies. BrightH α emission is seen from the head region along with some relatively fainter H α knots andH α arcs along two tails. UGCA 130
The BCD galaxy UGCA 130 is also a cometary type galaxy having a ‘head’ towards the southand a ‘tail’ extending towards the north with a bright source at the end. The H α emission is c (cid:13) , 1–, 1–
The BCD galaxy UGCA 130 is also a cometary type galaxy having a ‘head’ towards the southand a ‘tail’ extending towards the north with a bright source at the end. The H α emission is c (cid:13) , 1–, 1– ?? Jaiswal & Omar seen mainly in the head of the cometary type structure. Faint H α emission is also detectedfrom the bright source towards the end of the tail. L´opez-S´anchez (2010) speculated thatUGCA 130 is not undergoing any recent interaction but it has a disturbed H i kinematicsdue to some past interaction. NGC 2799
NGC 2799 is a member of close interacting galaxy pair with NGC 2798. It is an edge-ongalaxy with a prominent bar. A common stellar envelop fills the system indicating that twogalaxies are in an advanced stage of interaction leading to a possible merger. Both galaxiesshow prominent star formation in the H α images. NGC 2798 is classified as a starburstgalaxy (Kinney et al. 1993). Several extra-planner H α emitting knots are detected in thecommon stellar envelop. These H α knots may be tidal dwarf galaxies or H ii regions in thedisturbed stellar envelope caused by tidal interaction. MRK 22
MRK 22 is a blue compact galaxy having double nuclei (Mazzarella et al. 1991). A strongH α emission is seen from the bright nucleus. The double nuclei nature of this galaxy indicatesthat it is likely a merger of two dwarf systems (Pustilnik et al. 2001). A diffuse H α emissiontail is seen extending to the other nucleus. IC 2524
IC 2524 is classified as a dwarf disk galaxy. The disk is very faint. This galaxy has no visiblecompanion within 30 ′ radius. The H α emission is seen in the nuclear region along with weakdiffuse emission surrounding the nucleus. KUG 1013+381
KUG 1013+381 is a blue compact dwarf galaxy having intense nuclear star formation. Afaint small diffuse region is seen in r -band image at ∼ ′′ towards the south. Pustilnik &Martin (2007) speculated that the recent starburst phase in this galaxy has been triggeredthrough the interaction of its binary companion (UGC 5540) at an angular separation of ∼ ′ . The H α image shows bright nuclear region with two H α emission regions. NGC 3294 c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies α . The color of this region is significantly bluer than thegalaxy. This region has a measured recession velocity as 1464 km s − (Abazajian et al. 2009)compared to the galaxy’s recession velocity as 1586 km s − (de Vaucouleurs et al. 1991). Wespeculate that this WR H ii region is a dwarf galaxy undergoing a minor interaction withNGC 3294. NGC 3381
NGC 3381 is a spiral galaxy with two prominent spiral arms and a bright bar. The galaxyis highly lopsided in its stellar light distribution having diffuse envelope towards the west.The star formation is seen mainly in the nuclear region. Faint H α emission is also seen insmall patches along the spiral arms. NGC 3423
NGC 3423 is a face-on spiral galaxy with multiple spiral arms and a bright nucleus. A veryhigh concentration of star forming regions is seen in the disk. The H α region density ishigh in the central region. Ganda et al. (2007) noticed that the star-burst in this galaxy isvery recent on the basis of their absorption line strength analysis. Condon (1987) found adisturbed radio morphology at 1.4 GHz radio band. NGC 3430
The spiral galaxy NGC 3430 forms a pair with a star forming spiral galaxy NGC 3424 atabout 6 ′ (36 kpc) separation with 100 km s − velocity difference (Braine et al. 1993). Nord-gren et al. (1997) found a sign of tidal interaction in both the galaxies through optical andH i imaging. The star forming regions are seen in the nucleus and spiral arms. CGCG 038-051
CGCG 038-051 is a dwarf H ii galaxy of irregular shape. Three prominent star forming knotsare identified along the major axis of the galaxy in the H α image. The presence of multipledistinct nuclei within an asymmetric diffuse stellar envelope suggests that this system isundergoing a merger. c (cid:13) , 1– ?? Jaiswal & Omar
IC 2828
IC 2828 has an irregular morphology. The H α image shows multiple bright H α regions, tailsand plumes like features. Two bright H α regions are identified at the outer edges of thisgalaxy. The optical r -band image shows a faint tail extending towards the south. Thesefeatures are suggestive of a recent tidal interaction or a merger event. UM 439
UM 439 has an irregular morphology with two prominent nuclei. The H α image reveals mul-tiple nuclei. The tail and plume type features are also seen in the galaxy. Taylor et al. (1995)have found an asymmetric H i distribution, but they have not found any visible companion.These features are suggestive of a recent tidal interaction or a merger event. NGC 3949
NGC 3949 is a spiral galaxy with multiple spiral arms and a small bar. It is a member ofthe Ursa-Major cluster of galaxies. Maslowski & Kellermann (1988) mapped this galaxy at5 GHz radio band using VLA ‘B’ configuration and found that the location of the peakradio intensity has an offset of about 12 ′′ from the optical nucleus of the galaxy. Several starforming knots are identified in this galaxy, typical of a normal spiral galaxy. I SZ 059
Gil de Paz, Madore & Pevunova (2003) classified this object as nE BCD galaxy. This galaxyhas a very elongated elliptical envelope within which a clearly defined nucleus exists. Dou-blier, Caulet & Comte (1999) have shown using surface brightness distribution that it hasa disk. This galaxy is a member of the loose group NGC 4038 (Firth et al. 2006). StrongH α emission is seen from the nucleus. The r -band image reveals diffuse stellar envelop. CGCG 041-023
CGCG 041-023 is a barred spiral galaxy. Two symmetrically located prominent star formingregions are identified on the edges of the galaxy. This morphology is very intriguing. A weakH α emission is also seen along the bar. Further observations are required to understand theobserved morphology of this galaxy. SBS 1222+614 c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies ii galaxy. The diffuse H α emission is seen in the galaxy.A weak diffuse H α emitting arc is also visible in the south-east direction of the galaxy. NGC 4389
NGC 4389 is a barred spiral galaxy. Most of the star formation is concentrated along the bar.Some H α emitting knots are also seen in the disk. According to Sandage & Bedke (1994),the spiral arms of NGC 4389 are difficult to trace because of the large inclination angle. Afew isolated extra-planner H α emitting regions are detected in the halo of the galaxy. It islikely that this galaxy has undergone an interaction in the recent past. IC 3521
IC 3521 has an irregular morphology with a bar. This galaxy is a member of the Virgo clus-ter. Two prominent star forming knots are identified at the edges of the galaxy. The r -bandmorphology is patchy with a common diffuse stellar envelope. The stellar light distributionsuggests an ongoing merger. UGC 9273
UGC 09273 is classified as an irregular galaxy. The multiple star forming knots are detectedin a ring like structure, which is very intriguing. The stellar light distribution is lopsided.
MRK 475
MRK 475 is classified as a BCD galaxy. It shows strong nuclear star formation. The r -bandmorphology indicates lopsidedness in this galaxy. The estimated values of SFRs using the luminosities in different wave-bands are given inTable 7. The Figure 5 shows comparisons of the H α based SFR with the SFRs derived usingthe FUV, FIR and 1.4 GHz radio luminosities. In these plots, we have also included SFRestimates of WR galaxies provided in L´opez-S´anchez (2010) in order to enlarge our samplesize. The small and large spiral galaxies are designated by filled and open circles respectively. c (cid:13) , 1–, 1–
MRK 475 is classified as a BCD galaxy. It shows strong nuclear star formation. The r -bandmorphology indicates lopsidedness in this galaxy. The estimated values of SFRs using the luminosities in different wave-bands are given inTable 7. The Figure 5 shows comparisons of the H α based SFR with the SFRs derived usingthe FUV, FIR and 1.4 GHz radio luminosities. In these plots, we have also included SFRestimates of WR galaxies provided in L´opez-S´anchez (2010) in order to enlarge our samplesize. The small and large spiral galaxies are designated by filled and open circles respectively. c (cid:13) , 1–, 1– ?? Jaiswal & Omar
The error-weighted linear fits (solid line) to the data points were made. It can be seen fromFigure 5 that our sample is dominated by galaxies with SFR ≤ M ⊙ yr − , while majorityof galaxies in the L´opez-S´anchez (2010) sample have SFR ≥ M ⊙ yr − . These two samplesare therefore complimentary to each other. The H α -based SFR is seen well correlated withother SFR values estimated using FUV, FIR and 1.4 GHz radio luminosities in these plots.The strongest correlation is seen between H α and FUV SFRs. The slope of the H α -FUVcorrelation is close to unity. This result is in agreement with previous studies (e.g., Buat,Donas & Deharveng 1987; Buat 1992; Sullivan et al. 2000; Bell and Kennicutt 2001; L´opez-S´anchez 2010). The slopes of the FIR-H α and radio-H α SFR relations are ∼ . ∼ . α and FIR-H α are almostidentical at ∼ . ∼ .
4. The radio-H α correlation shows larger scatter compared toother correlations.Various interplay between the calibration constants and age, IMF, and other physicalprocesses in ISM are described in Calzetti (2012). The tight correlation between FUV andH α SFR is not surprising. The H α based SFR estimates trace the youngest star formation.The FUV photons from the most massive stars ( M ≥ M ⊙ ) are responsible for the H α line.The ionizing FUV flux decreases by two orders of magnitude between 5 Myr and 10 Myrafter the burst. While the massive O-type stars die in a few tens of Myr time scale, theout-numbered B-type stars continue contributing to the total FUV flux over longer timescales. On shorter time scales of a few Myr, the total FUV flux emitted by O and B typestars collectively can vary up to a factor of 3. The calibration constants for convertingluminosities to SFR in different bands are only an approximation. The best accuracies inthese calibrations are ± α band. It also appears from Figure 5 that scatteris slightly larger towards the low SFR regime, which is dominated by small-size galaxies. Thelog-log rms scatter values between 0.4 and 0.5 in these correlations can be understood withinvarious uncertainties and inaccuracies in calibrations, particularly for WR galaxies wherestar-burst is very young. Our results are largely consistent with previous studies in termsof scatter in various SFR correlations (Bell and Kennicutt 2001; Lee et al. 2009; Hao et al.2011). In small galaxies, the most recent single burst dominated by WR region may oftenbe representing the total star formation. On the other hand, in large galaxies, episodic starformation spread over longer time scales (a few 100 Myr) may be present. This conjecture issupported by the color composite SDSS and H α images, which show that large galaxies have c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies α SFRs. TheFIR luminosity was determined using 60 µ m and 100 µ m flux densities. The slope of the FIR-H α SFR correlation is significantly deviating from unity with an indication that FIR fluxesare under-estimating SFR in the lower SFR sides. The FIR emission is thermal emissionfrom the dust heated by FUV photons, emitted by hot and young stars. The FIR luminositydepends also on the dust content and its distribution relative to star forming regions. Thecalibration for FIR luminosity to SFR is largely based on large galaxies where older starsalso contribute significantly to the heating of dust. We speculate that in low SFR range,star-light spectral energy distribution (SED) plays a more deciding role for 60 µ m and100 µ m fluxes. A total (bolometric) infrared luminosity may give better estimates for SFR(Calzetti 2013). Finally, the radio and H α SFRs are also correlated, but with the largestscatter. The radio emission is primarily of synchrotron (non-thermal) origin in star forminggalaxies, with relatively small contribution from thermal free-free emission from the ionizedgas. The slope of the radio-H α SFR correlation indicates that towards high SFR ends, radioemission is consistent with that expected from the H α emission. However, radio emission isunder-estimating SFR towards the low SFR range. The scatter is also significantly highertowards the low SFR range. The non-thermal radio continuum emission is expected to lastfor longer duration, nearly 100 Myr after the initial starburst, compared to the H α emission.The episodic star-bursts separated by a few tens of Myr in a galaxy will further complicatea comparison between these tracers of SFR. There is a possibility that young supernovaeresponsible for accelerating cosmic electrons to the relativistic speeds are lacking in the veryyoung starbursts in WR galaxies. Several dwarf galaxies in our sample are not detected inthe radio continuum. The radio emission from galaxies also depends on the strengths ofmagnetic field in galaxies. The magnetic field in dwarf galaxies is still poorly understood(Beck & Wielebinski 2013).In order to further understand radio emission from WR galaxies, we estimated expected c (cid:13) , 1– ?? Jaiswal & Omar thermal radio continuum from the knowledge of the H α flux in these galaxies, using therelation given by Dopita et al. (2002), F . GHz ( thermal ) [mJy] = 1 . × F H α [erg cm − s − ] (2)The thermal radio continuum is subtracted from the total radio emission to get anestimate for non-thermal radio continuum. The ratio (R) of non-thermal to thermal radioflux is given in Table 8. The average value of this ratio for a sample of star-burst galaxieswas estimated as log R = 1 . ± . α ). This plot indicates a weak positive correlation betweenthese two quantities. The higher values of FUV to H α ratio are expected in relatively olderstar-bursts as the H α emission decreases faster than total FUV flux in the first tens ofMyr from the initial burst time. In older star-bursts, supernovae explosions are expected togive rise to normal non-thermal radio flux. Therefore, this hint of correlation between ‘R’and SFR(FUV)/SFR(H α ) supports our initial observation that majority of WR galaxies arelacking non-thermal radio flux. A tight correlation between radio continuum and FIR emission is known in normal starforming galaxies (e.g., Helou et al. 1985; Condon 1992; Niklas & Beck 1997; Yun et al. 2001).The radio-FIR correlation in star-forming galaxies is understood as follows. The dust absorbsFUV photons produced by hot stars, and thereby gets heated. This warm dust re-radiatesin the FIR waveband, producing a linear correlation between SFR and FIR luminosity. TheGHz radio continuum emission from star-forming galaxies is mainly synchrotron radiationfrom the cosmic electrons accelerated to relativistic speeds through supernova explosionsof the young massive stars. Therefore, a radio-FIR correlation is expected in star forminggalaxies. The tightness of the radio-FIR correlation is estimated in terms of a parameter ( q ),which is defined as the logarithmic of the ratio of total FIR flux density and radio continuumflux density at 1.4 GHz (see, Helou et al. 1985). q = log (cid:18) . S µ m + S µ m . S . GHz (cid:19) (3)where all flux densities are in Jy. The average value of q for normal galaxies has beenestimated as < q > ∼ . ± .
26 (Yun et al. 2001). The most remarkable feature of the radio- c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies ∼ q values arefalling within the scatter seen for normal galaxies. However, we notice a possible trend thatgalaxies at low FIR or low SFR end, are showing higher value of q . These trends can againbe understood as a general deficiency of non-thermal radio flux in WR galaxies as discussedin the previous section. It is worth to point out that some WR galaxies such as MRK 996(Jaiswal & Omar 2013), and several galaxies in L´opez-S´anchez (2010) sample show N/Oenrichment. The N/O enrichment is expected to take place in young (WR) star-burst regions,where nitrogen is enriching the ISM through the stellar winds, however, similar oxygenenrichment expected from supernovae explosions is lacking (Kobulnicky et al. 1997; Putilniket al. 2004; Brinchmann et al. 2008; L´opez-S´anchez & Esteban et al. 2010b). Although,magnetic field strengths in dwarf galaxies are not known very well, a few radio polarimetricobservations of dwarf galaxies reveal presence of magnetic fields in these systems. (Chy˙zy etal. 2000; Gaensler et al. 2005; Chy˙zy et al. 2011; Beck & Wielebinski 2013; Drzazga et al.2016). Our analysis indicates that radio deficiency in several WR galaxies is mainly due tolack of supernovae events. The optical morphologies of WR galaxies in our sample are studied using the SDSS andH α images. The color composite SDSS images reveal presence of blue star-forming disks inspiral galaxies. Most of the galaxies show very bright distinct blue regions. The fiber-basedSDSS spectrum were taken mainly on these bright blue regions, which showed WR features.The H α emission corresponding to the blue regions is detected in all the galaxies. The H α andoptical r -band morphologies of all galaxies in the sample were visually checked for signa-tures of tidal interactions. L´opez-S´anchez (2010) identified high occurrence of interactionrelated morphological features (plume, tails, merger and tidal dwarf galaxies), kinematicaldisturbances, and differences in chemical compositions of H ii regions in their sample of WR c (cid:13) , 1– ?? Jaiswal & Omar galaxies. The presence of multiple nuclei, arcs and tidal tails are generally considered as asignature of recent or ongoing tidal interaction (e.g., Beck & Kovo 1999; L´opez-S´anchez et al.2004a; Matsui et al. 2012; Adamo et al. 2012). A summary of the prominent morphologicalfeatures seen in WR galaxies in our sample is provided in Table 9. The interaction prob-ability in this table is inferred based on various morphological features. Galaxies showingH i or optical tidal tails and distinct multiple nuclei with a disturbed optical envelope aretermed as highly probable. Galaxies showing lopsidedness or asymmetric light distributionalong with any other feature such as probable multiple nuclei or arcs are termed as moder-ately probable. Galaxies without any significant disturbed optical morphology are labeledwith low interaction probability. It should be noted that several galaxies termed as havinglow probability show some interaction features such as bar, mild lopsidedness, or nuclearstar formation. However, in absence of any strong tidal feature, we preferred to label theseobjects with low interaction probability.In our sample, we find that 5 galaxies show prominent tidal tails or cometary shapes, 7galaxies have bars, and 7 galaxies have two or three bright dominating H α regions suspectedas multiple nuclei. The star formation dominated in the nuclear region is seen in 8 galaxies. Inaddition, two galaxies show presence of extra-planner H α regions. One galaxy has misalignedH α bar with respect to the optical bar. A large number of 16 galaxies show lopsidedness(stellar light distribution or star formation) or irregular morphologies in our sample. Threegalaxies show lopsided or disturbed radio continuum emission. Two galaxies are known tohave disturbed H i morphologies. It is not clear if multiple bright dominating H α regions arejust different H ii regions or multiple nuclei of different galaxies such as other low mass dwarfgalaxies in a merger stage. Kinematical and chemical abundance measurements will be usefulto distinguish these two scenarios. The lopsidedness in galaxies has been studied previouslyin other galaxies, and is believed to be caused by tidal interactions (e.g., Jog 1997; Zaritsky& Rix 1997; Angiras et al. 2006). The stellar bars can also result from recent tidal interactionevents (e.g., Barnes & Hernquist 1991; Miwa & Noguchi 1998). The dwarf systems in oursample often show intense nuclear star-burst consistent with previous results (e.g., Strickland& Stevens 1999; Adamo et al. 2011). It has been seen in N-body simulations (e.g., Hernquist& Mihos 1995) that tidal interaction and mergers between galaxies can channel gas towardsthe center of galaxy as a result of loss of angular momentum. This gas can give rise tonuclear starburst in galaxies (Mihos & Hernquist 1994). The optical morphological featuresseen in our study are also highly suggestive of prevalence of tidal interactions and mergers c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies i i observations will help in understanding tidal interactionsin these galaxies to greater depths. Twenty-five Wolf-Rayet galaxies were imaged in the H α emission-line and r -band using1.3-meter and 2-meter optical telescopes. These galaxies were selected from the SDSS withthe distance cutoff at ∼
25 Mpc. Majority of the selected galaxies in our sample are ofsmall-size ( ≤
10 kpc) or dwarf galaxies. The H α sensitivity obtained in our observations istypically 10 − erg s − cm − arcsec − . The integrated H α fluxes were corrected for internaland Galactic extinction, and also for the line contaminations in the H α and r passbands.The star formation rates were derived for all the galaxies using H α , FUV, FIR and radioluminosities. The archival data in the FUV, FIR and 1.4 GHz radio bands were takenfrom GALEX, IRAS and VLA surveys (FIRST and NVSS). We combined our sample withthe sample of 20 WR galaxies studied by L´opez-S´anchez (2010). These two samples arecomplimentary to each other in the sense that our sample is dominated by galaxies withSFR ≤ M ⊙ yr − while majority of galaxies in the L´opez-S´anchez (2010) sample has higherSFR values.The SFRs estimated from the luminosities in different wavebands were compared with theH α based SFR. The H α based SFR are found well correlated with other SFRs. We noticed,in general, that scatter in the correlations is higher towards the low SFR range. All theslopes in correlation with the H α SFR deviate from the unity. The highest deviation is seenwith radio SFRs and the smallest deviation is with FUV based SFRs. These deviations andscatter are expected to be resulting mainly from uncertainties in the calibration constantsfor SFR-luminosity relations, star formation histories, age of the most recent star-burst, and c (cid:13) , 1– ?? Jaiswal & Omar assumptions about various physical processes in the ISM. These results are consistent, ingeneral, with other similar studies. The radio-FIR correlation further reveals that majority ofWR galaxies towards low SFR range are radio deficient, most likely due to lack of supernovaeevents in the young star-bursts in the WR galaxies. This result is also supported by N/Oenrichment seen in several WR galaxies.The optical morphologies reveal presence of active star formation, wide-spread in thedisks of spiral galaxies in our sample. All WR galaxies are dominated by a distinct region ofvery blue star forming region, coincident with the WR features detected in SDSS. Severalsmall-size and dwarf galaxies in our sample show multiple H α emission region. It is notobvious if these multiple regions are just H ii regions or nuclei of separate dwarf galaxiesin the merger state. Several galaxies in our sample show tidal features and lopsidedness intheir stellar light distribution. Bars and nuclear star-bursts are also seen in a few galaxies.The morphologies of a total of 16 galaxies are inferred here to be suggestive of tidallyinteracting. In general, our study reveals tidal interactions as the prime reason for starformation trigger in WR galaxies, consistent with previous studies on WR galaxies (M´endez& Esteban 2000; L´opez-S´anchez & Esteban 2008; L´opez-S´anchez 2010). Overall about two-thirds of WR galaxies in different samples show some signatures of tidal interactions ormergers. Further studies using the data presented in this paper are being carried out, inorder to understand various SFR tracers in more details with an emphasis to effects ofvery young star-burst in WR galaxies. The H i α emitting regions in order to reveal details of tidal interactionsin these galaxies. ACKNOWLEDGMENTS
We thank the referee, A. R. L´opez-S´anchez for critically examining the manuscript, that greatly improved the clarity andcontents of this paper. IRAF (Image Reduction and Analysis facility) is distributed by NOAO which is operated by AURAInc., under cooperative agreement with NSF. This research has made use of the NASA/IPAC Extragalactic Database (NED)which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the NationalAeronautics and Space Administration. This research has made use of NASA’s Astrophysics Data System. We thank thestaff of ARIES, whose dedicated efforts made these observations possible. DFOT is run by Aryabhatta Research Institute ofObservational Sciences with support from the Department of Science and Technology, Govt. of India. We wish to acknowledgethe IUCAA/IGO staff for their support during our observations. We acknowledge the use of the SDSS. Funding for the SDSS andSDSS-II has been provided by the Alfred P. Sloan Foundation, the Participating Institutions, the National Science Foundation,c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies REFERENCES
Abazajian K. N., et al., 2009, ApJS, 182, 543Adamo A., ¨Ostlin G., Zackrisson E., Papaderos P., Bergvall N., Rich R. M., Micheva G.,2011, MNRAS, 415, 2388Adamo A., et al., 2012, MNRAS, 426, 1185Allen D. A., Wright A. E., Goss W. M., 1976, MNRAS, 177, 91Amorisco N. C., Evans N. W., van de Ven G., 2014, Natur, 507, 335Angiras R. A., Jog C. J., Omar A., Dwarakanath K. S., 2006, MNRAS, 369, 1849Ashley T. L., 2014, PhDT, 198Asplund M., Grevesse N., Sauval A. J., Scott P., 2009, ARA&A, 47, 481Baldwin J. A., Spinrad H., Terlevich R., 1982, MNRAS, 198, 535Barnes J. E., Hernquist L. E., 1991, ApJ, 370, L65Barnes D. G., et al., 2001, MNRAS, 322, 486Beck S. C., Kovo O., 1999, AJ, 117, 190Beck R., Wielebinski R., 2013, pss5.book, 5, 641Becker R. H., White R. L., Helfand D. J., 1995, ApJ, 450, 559Bekki K., 2008, MNRAS, 388, L10Bell E. F., Kennicutt R. C., Jr., 2001, ApJ, 548, 681Binggeli B., Popescu C. C., Tammann G. A., 1993, A&AS, 98, 275 c (cid:13) , 1– ?? Jaiswal & Omar
Blackman C. P., 1980, MNRAS, 190, 459Braine J., Combes F., Casoli F., Dupraz C., Gerin M., Klein U., Wielebinski R., BrouilletN., 1993, A&AS, 97, 887Bravo-Alfaro H., Brinks E., Baker A. J., Walter F., Kunth D., 2004, AJ, 127, 264Brinchmann J., Kunth D., Durret F., 2008, A&A, 485, 657Brinks E., Klein U., 1988, MNRAS, 231, 63PBrosch N., Almoznino E., Heller A. B., 2004, MNRAS, 349, 357Buat V., Donas J., Deharveng J. M., 1987, A&A, 185, 33Buat V., 1992, A&A, 264, 444Buat V., Boselli A., Gavazzi G., Bonfanti C., 2002, A&A, 383, 801Cair´os L. M., Caon N., V´ılchez J. M., Gonz´alez-P´erez J. N., Mu˜noz-Tu˜n´on C., 2001, ApJS,136, 393Calzetti D., 1997, AIPC, 408, 403Calzetti D., et al., 2007, ApJ, 666, 870Calzetti, D., Wu, S.-Y., Hong, S., et al., 2010, ApJ, 714, 1256Calzetti D., 2011, EAS, 46, 133Calzetti D., 2013, seg..book, 419Cardelli J. A., Clayton G. C., Mathis J. S., 1989, ApJ, 345, 245Chy˙zy K. T., Beck R., Kohle S., Klein U., Urbanik M., 2000, A&A, 355, 128Chy˙zy K. T., We˙zgowiec M., Beck R., Bomans D. J., 2011, A&A, 529, A94Comte G., Petrosian A. R., Ohanian G. A., Stepanian J. A., 1999, Ap, 42, 149Condon J. J., 1987, ApJS, 65, 485Condon J. J., 1992, ARA&A, 30, 575Condon J. J., Cotton W. D., Greisen E. W., Yin Q. F., Perley R. A., Taylor G. B., BroderickJ. J., 1998, AJ, 115, 1693Condon J. J., Cotton W. D., Broderick J. J., 2002, AJ, 124, 675Conti P. S., 1991, ApJ, 377, 115Cram L., Hopkins A., Mobasher B., Rowan-Robinson M., 1998, ApJ, 507, 155Das H. K., Menon S. M., Paranjpye A., Tandon S. N., 1999, BASI, 27, 609Deeg H.-J., Duric N., Brinks E., 1997, A&A, 323, 323de Vaucouleurs G., de Vaucouleurs A., Corwin J. R., 1976, RC2, 0de Vaucouleurs G., de Vaucouleurs A., Corwin H. G., Jr., Buta R. J., Paturel G., Fouqu´eP., 1991, rc3..book, I, c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies c (cid:13) , 1–, 1–
Blackman C. P., 1980, MNRAS, 190, 459Braine J., Combes F., Casoli F., Dupraz C., Gerin M., Klein U., Wielebinski R., BrouilletN., 1993, A&AS, 97, 887Bravo-Alfaro H., Brinks E., Baker A. J., Walter F., Kunth D., 2004, AJ, 127, 264Brinchmann J., Kunth D., Durret F., 2008, A&A, 485, 657Brinks E., Klein U., 1988, MNRAS, 231, 63PBrosch N., Almoznino E., Heller A. B., 2004, MNRAS, 349, 357Buat V., Donas J., Deharveng J. M., 1987, A&A, 185, 33Buat V., 1992, A&A, 264, 444Buat V., Boselli A., Gavazzi G., Bonfanti C., 2002, A&A, 383, 801Cair´os L. M., Caon N., V´ılchez J. M., Gonz´alez-P´erez J. N., Mu˜noz-Tu˜n´on C., 2001, ApJS,136, 393Calzetti D., 1997, AIPC, 408, 403Calzetti D., et al., 2007, ApJ, 666, 870Calzetti, D., Wu, S.-Y., Hong, S., et al., 2010, ApJ, 714, 1256Calzetti D., 2011, EAS, 46, 133Calzetti D., 2013, seg..book, 419Cardelli J. A., Clayton G. C., Mathis J. S., 1989, ApJ, 345, 245Chy˙zy K. T., Beck R., Kohle S., Klein U., Urbanik M., 2000, A&A, 355, 128Chy˙zy K. T., We˙zgowiec M., Beck R., Bomans D. J., 2011, A&A, 529, A94Comte G., Petrosian A. R., Ohanian G. A., Stepanian J. A., 1999, Ap, 42, 149Condon J. J., 1987, ApJS, 65, 485Condon J. J., 1992, ARA&A, 30, 575Condon J. J., Cotton W. D., Greisen E. W., Yin Q. F., Perley R. A., Taylor G. B., BroderickJ. J., 1998, AJ, 115, 1693Condon J. J., Cotton W. D., Broderick J. J., 2002, AJ, 124, 675Conti P. S., 1991, ApJ, 377, 115Cram L., Hopkins A., Mobasher B., Rowan-Robinson M., 1998, ApJ, 507, 155Das H. K., Menon S. M., Paranjpye A., Tandon S. N., 1999, BASI, 27, 609Deeg H.-J., Duric N., Brinks E., 1997, A&A, 323, 323de Vaucouleurs G., de Vaucouleurs A., Corwin J. R., 1976, RC2, 0de Vaucouleurs G., de Vaucouleurs A., Corwin H. G., Jr., Buta R. J., Paturel G., Fouqu´eP., 1991, rc3..book, I, c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies c (cid:13) , 1–, 1– ?? Jaiswal & Omar
Jog C. J., 1997, ApJ, 488, 642Joseph R. D., Wright G. S., 1985, MNRAS, 214, 87Karthick M. C., L´opez-S´anchez ´A. R., Sahu D. K., Sanwal B. B., Bisht S., 2014, MNRAS,439, 157Kennicutt R. C., Jr., Kent S. M., 1983, AJ, 88, 1094Kennicutt R. C., Jr., Roettiger K. A., Keel W. C., van der Hulst J. M., Hummel E., 1987,AJ, 93, 1011Kennicutt R. C., Jr., 1998a, ARA&A, 36, 189Kennicutt R. C., Jr., 1998b, ApJ, 498, 541Kennicutt R. C., Jr., Lee J. C., Funes S. J., Jos´e G., Sakai S., Akiyama S., 2008, ApJS,178, 247Kinney A. L., Bohlin R. C., Calzetti D., Panagia N., Wyse R. F. G., 1993, ApJS, 86, 5Kobulnicky H. A., Skillman E. D., Roy J.-R., Walsh J. R., Rosa M. R., 1997, ApJ, 477,679Koribalski B., 1996, ASPC, 106, 238Koribalski B. S., Staveley-Smith L., Kilborn V. A., et al., 2004, AJ, 128, 16Kunth D., Joubert M., 1985, A&A, 142, 411Kunth D., Sargent W. L. W., 1981, A&A, 101, L5Kunth D., Schild H., 1986, A&A, 169, 71Larson R. B., Tinsley B. M., 1978, ApJ, 219, 46Lee J. C., et al., 2009, ApJ, 706, 599Lelli F., Verheijen M., Fraternali F., 2014, MNRAS, 445, 1694Leroy A., Bolatto A. D., Simon J. D., Blitz L., 2005, ApJ, 625, 763Lisenfeld U., Voelk H. J., Xu C., 1996, A&A, 314, 745L´opez-S´anchez ´A. R., Esteban C., Rodr´ıguez M., 2004a, ApJS, 153, 243L´opez-S´anchez ´A. R., Esteban C., Rodr´ıguez M., 2004b, A&A, 428, 425L´opez-S´anchez ´A. R., Esteban C., Garc´ıa-Rojas J., 2006, A&A, 449, 997L´opez-S´anchez ´A. R., Esteban C., 2008, A&A, 491, 131L´opez-S´anchez A. R., Esteban C., 2009, A&A, 508, 615L´opez-S´anchez ´A. R., Esteban C., 2010a, A&A, 516, A104L´opez-S´anchez ´A. R., Esteban C., 2010b, A&A, 517, A85L´opez-S´anchez ´A. R., 2010, A&A, 521, A63L´opez-S´anchez ´A. R., Koribalski B. S., van Eymeren J., Esteban C., Kirby E., Jerjen H., c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies c (cid:13) , 1–, 1–
Jog C. J., 1997, ApJ, 488, 642Joseph R. D., Wright G. S., 1985, MNRAS, 214, 87Karthick M. C., L´opez-S´anchez ´A. R., Sahu D. K., Sanwal B. B., Bisht S., 2014, MNRAS,439, 157Kennicutt R. C., Jr., Kent S. M., 1983, AJ, 88, 1094Kennicutt R. C., Jr., Roettiger K. A., Keel W. C., van der Hulst J. M., Hummel E., 1987,AJ, 93, 1011Kennicutt R. C., Jr., 1998a, ARA&A, 36, 189Kennicutt R. C., Jr., 1998b, ApJ, 498, 541Kennicutt R. C., Jr., Lee J. C., Funes S. J., Jos´e G., Sakai S., Akiyama S., 2008, ApJS,178, 247Kinney A. L., Bohlin R. C., Calzetti D., Panagia N., Wyse R. F. G., 1993, ApJS, 86, 5Kobulnicky H. A., Skillman E. D., Roy J.-R., Walsh J. R., Rosa M. R., 1997, ApJ, 477,679Koribalski B., 1996, ASPC, 106, 238Koribalski B. S., Staveley-Smith L., Kilborn V. A., et al., 2004, AJ, 128, 16Kunth D., Joubert M., 1985, A&A, 142, 411Kunth D., Sargent W. L. W., 1981, A&A, 101, L5Kunth D., Schild H., 1986, A&A, 169, 71Larson R. B., Tinsley B. M., 1978, ApJ, 219, 46Lee J. C., et al., 2009, ApJ, 706, 599Lelli F., Verheijen M., Fraternali F., 2014, MNRAS, 445, 1694Leroy A., Bolatto A. D., Simon J. D., Blitz L., 2005, ApJ, 625, 763Lisenfeld U., Voelk H. J., Xu C., 1996, A&A, 314, 745L´opez-S´anchez ´A. R., Esteban C., Rodr´ıguez M., 2004a, ApJS, 153, 243L´opez-S´anchez ´A. R., Esteban C., Rodr´ıguez M., 2004b, A&A, 428, 425L´opez-S´anchez ´A. R., Esteban C., Garc´ıa-Rojas J., 2006, A&A, 449, 997L´opez-S´anchez ´A. R., Esteban C., 2008, A&A, 491, 131L´opez-S´anchez A. R., Esteban C., 2009, A&A, 508, 615L´opez-S´anchez ´A. R., Esteban C., 2010a, A&A, 516, A104L´opez-S´anchez ´A. R., Esteban C., 2010b, A&A, 517, A85L´opez-S´anchez ´A. R., 2010, A&A, 521, A63L´opez-S´anchez ´A. R., Koribalski B. S., van Eymeren J., Esteban C., Kirby E., Jerjen H., c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies c (cid:13) , 1–, 1– ?? Jaiswal & Omar
University Science Books, Mill Valley, CAPrabhu T. P., 2012, ASInC, 4, 131Price R., Duric N., 1992, ApJ, 401, 81Pustilnik S. A., Kniazev A. Y., Lipovetsky V. A., Ugryumov A. V., 2001, A&A, 373, 24Pustilnik S., Kniazev A., Pramskij A., Izotov Y., Foltz C., Brosch N., Martin J.-M., Ugryu-mov A., 2004, A&A, 419, 469Pustilnik S. A., Martin J.-M., 2007, A&A, 464, 859Ranalli P., Comastri A., Setti G., 2003, A&A, 399, 39Rosa-Gonz´alez D., Terlevich E., Terlevich R., 2002, MNRAS, 332, 283Roussel H., Sauvage M., Vigroux L., Bosma A., 2001, A&A, 372, 427Sagar R., Omar A., Kumar B., Gopinathan M., Pandey S. B., Bangia T., Pant J. S., ShuklaV., Yadava S., 2011, Current Science, 101, 1020Sage L. J., Salzer J. J., Loose H.-H., Henkel C., 1992, A&A, 265, 19Salim S., et al., 2007, ApJS, 173, 267Sandage A., Bedke J., 1994, cag..book, 638Sanders D. B., Mirabel I. F., 1996, ARA&A, 34, 749Sanders D. B., Mazzarella J. M., Kim D.-C., Surace J. A., Soifer B. T., 2003, AJ, 126, 1607Schaerer D., Contini T., Pindao M., 1999, A&AS, 136, 35Schaerer D., Vacca W. D., 1998, ApJ, 497, 618Schlafly E. F., Finkbeiner D. P., 2011, ApJ, 737, 103Schlegel D. J., Finkbeiner D. P., Davis M., 1998, ApJ, 500, 525Schneider S. E., Thuan T. X., Magri C., Wadiak J. E., 1990, ApJS, 72, 245Shlosman I., 2013, seg..book, 555Smoker J. V., Davies R. D., Axon D. J., Hummel E., 2000, A&A, 361, 19Soifer B. T., Boehmer L., Neugebauer G., Sanders D. B., 1989, AJ, 98, 766Solomon P. M., Sage L. J., 1988, ApJ, 334, 613Spector O., Finkelman I., Brosch N., 2012, MNRAS, 419, 2156Strickland D. K., Stevens I. R., 1999, MNRAS, 306, 43Surace J. A., Sanders D. B., Mazzarella J. M., 2004, AJ, 127, 3235Sullivan M., Treyer M. A., Ellis R. S., Bridges T. J., Milliard B., Donas J., 2000, MNRAS,312, 442Taylor C. L., Brinks E., Grashuis R. M., Skillman E. D., 1995, ApJS, 99, 427Thuan T. X., Lipovetsky V. A., Martin J.-M., Pustilnik S. A., 1999, A&AS, 139, 1 c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies APPENDIX A: STAR FORMATION RATE CALCULATIONS
A brief overview of methods used to convert luminosities at different wavelengths to SFR isprovided here.
A1: Far-ultraviolet emission
The ultraviolet (UV) radiation is emitted by the most massive stars with their surfacetemperatures in the range of T ∼ − λ ≤
912 ˚A) is almost completely absorbed by the surrounding interstellarmedium (ISM). However, the radiation longward of this limit ( λ >
912 ˚A) can escape theISM and provides an estimate for the total UV radiation from the massive stars. The FUV(1344-1786 ˚A) band continuum emission can therefore be used to measure SFR in galaxies(e.g., Salim et al. 2007).The FUV flux density is estimated here from the GALEX data. This flux density iscorrected for foreground and internal extinctions. The GALEX FUV luminosity is convertedto SFR using the relation given by Kennicutt (1998a):
SF R
F UV [ M ⊙ yr − ] = L F UV [erg s − ˚A − ]9 . × (4) A2: H α emission The absorbed FUV radiation shortward of the Lyman continuum limit ionizes the ISM,which in turn emits nebular emission lines. The strengths of nebular emission lines are c (cid:13) , 1– ?? Jaiswal & Omar directly proportional to the impinging FUV flux. The nebular lines therefore can provide anindirect measurement of the FUV flux from the most massive stars and hence in turn anestimate for the SFR. The H α line ( λ = 6563 ˚A), being the strongest among the nebularemission lines from H ii regions, is widely used to trace recent star formation in galaxies(e.g., Kennicutt 1998a). The SFR Hα is estimated here from the H α luminosity using therelation given Kennicutt (1998b) : SF R Hα [ M ⊙ yr − ] = L Hα [erg s − ]1 . × (5) A3: Far-infrared emission
The FUV radiation also heats the dust in the ISM. The dust emits thermal radiation in thefar infrared bands (40 − µ m: Devereux & Young 1990; Devereux & Hameed 1997). TheFIR emission therefore also indirectly traces the star formation rates in galaxies. The IRASdata provides the 60 µ m and 100 µ m flux densities, which can be converted into total FIRluminosity using the relations described in Yun et al. (2001):log L µ m [ L ⊙ ] = 6 .
014 + 2 log D + log S µ m (6) L F IR [ L ⊙ ] = (cid:18) S µ m . S µ m (cid:19) L µ m [ L ⊙ ] (7)where D is the distance in Mpc and S µ m and S µ m are flux densities at 60 µ m and 100 µ m,respectively in units of Jy. The SFR F IR can be then estimated using the relation given byKennicutt (1998b) :
SF R
F IR [ M ⊙ yr − ] = L F IR [ L ⊙ ]5 . × (8) A4: 1.4 GHz radio emission
The most massive stars ( M ≥ M ⊙ ) end their life in supernovae explosions within about 30Myr of their formation. These powerful explosions accelerate cosmic electrons to relativisticspeeds. These electrons in presence of galactic magnetic field emit synchrotron (non-thermal)radiation. This radiation follows a power law in frequency where spectral power at lowfrequencies is higher than that at higher frequencies. The NVSS and FIRST are the two verysensitive radio surveys carried out at 1.4 GHz using the VLA. Most of the understandingon radio emission from nearby galaxies is due to these radio surveys. The 1.4 GHz radioemission from normal star forming galaxies is dominated by the synchrotron emission. Thefree-free thermal emission from star forming regions also contributes at a level of ∼ c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies ∼
100 Myr, i.e., typical lifetimes of relativistic cosmic electrons in galaxies.The 1.4 GHz radio continuum flux density obtained from NVSS and FIRST is convertedinto radio luminosity using the relation given by Yun et al. (2001):log L . [W Hz − ] = 20 .
08 + 2 log D + log S . (9)where D is the distance in Mpc and S . GHz is the 1.4 GHz radio continuum flux density inJy. The SFR . is then estimated using the relation given by Condon et al. (2002): SF R . GHz [ M ⊙ yr − ] = L . GHz [W Hz − ]4 . × (10) c (cid:13) , 1– ?? J a i s w a l &O m a r Table 1.
Basic properties of WR galaxies in our sample.Name RA (J2000) DEC (J2000) Type r v helio
Size Other Namesh m s ◦ ′ ′′ [mag] [km/s] [arcmin]UM 311 01 15 34.4 −
00 51 46 BCD 17 .
83 1675 ± .
05 1535 ± −
01 09 06 SAB(rs)c 12 .
91 1608 ± a −
00 29 55 SAB(rs)c 11 .
69 1517 ± b . A ± c . A ± c .
63 1673 ± d .
67 1551 ± e .
33 1450 ± .
00 1173 ± .
54 1586 ± c .
84 1629 ± f .
93 1011 ± a .
14 1586 ± g .
04 1021 ± .
71 1039 ± h . A ± i .
22 800 ± j −
19 37 27 S0 14 . A ± k .
73 1350 ± l .
70 706 ± e .
07 718 ± j .
24 595 ± m .
03 1289 ± n .
30 583 ± e v helio : a HI Parkes All Sky Survey (HIPASS: Barnes et al. 2001), b Koribalski et al. (2004), c Third Reference Catalogue (RC3: de Vaucouleurs et al. 1991), d Monnier Ragaigne et al. (2003), e Thuan et al. (1999), f van Driel et al. (2001), g Nordgren et al. (1997), h Smoker et al. (2000), i Comte et al. (1999), j Verheijen & Sancisi (2001), k Firth et al. (2006), l Lu et al. (1993), m Binggeli et al. (1993), n Schneider et al. (1990). The v helio values for other galaxies are taken from SDSS catalogues (Abazajian et al. 2009).The r -band magnitudes super-scripted by A are calculated using Johnson B and Cousins R band magnitudes from Gil de Paz et al. (2003), and Lupton transformation equations. Themagnitudes for other galaxies are taken from the SDSS (Abazajian et al. 2009). c (cid:13) R A S , M N R A S , ?? α imaging survey of Wolf-Rayet galaxies Table 2.
Extinction parameters and metallicities.Galaxy name f Hα /f Hβ E g ( B − V ) E f ( B − V ) 12 + log(O/H)[mag] [mag]UM 311 3 . ± .
09 0 . ± .
03 0 .
03 8 . ± . f IC 225 3 . ± .
06 0 . ± .
02 0 .
03 8 . ± . . ± .
10 0 . ± .
03 0 .
03 8 . ± . . ± .
15 0 . ± .
04 0 .
03 8 . ± . . ± . a — 0 .
72 8 . ± . g UGCA 130 4 . ± . b — 0 .
07 8 . ± . h NGC 2799 3 . ± .
11 0 . ± .
03 0 .
02 8 . ± . . ± .
13 0 . ± .
04 0 .
01 8 . ± . i IC 2524 3 . ± .
08 0 . ± .
02 0 .
01 8 . ± . . ± . c — 0 .
01 7 . ± . . ± .
40 0 . ± .
10 0 .
02 8 . ± . . ± .
09 0 . ± .
03 0 .
02 8 . ± . . ± .
37 0 . ± .
12 0 .
03 8 . ± . . ± .
16 0 . ± .
04 0 .
02 8 . ± . . ± .
10 0 . ± .
03 0 .
03 7 . ± . . ± .
15 0 . ± .
04 0 .
05 8 . ± . . ± .
08 0 . ± .
03 0 .
02 8 . ± . j NGC 3949 3 . ± .
09 0 . ± .
02 0 .
02 8 . ± .
02I SZ 59 4 . ± . d — 0 . ∼ . k CGCG 041-023 3 . ± .
31 0 . ± .
09 0 .
01 8 . ± . . ± .
08 0 . ± .
03 0 .
01 8 . ± . l NGC 4389 3 . ± .
10 0 . ± .
03 0 .
01 8 . ± . . ± .
42 0 . ± .
11 0 .
02 8 . ± . . ± .
08 0 . ± .
03 0 .
02 8 . ± . . ± . e — 0 .
01 7 . ± . ma f Hα /f Hβ = 6 . ± . E ( B − V ) = 0 . ± .
01 (Guseva, Izotov & Thuan 2000). b f Hα /f Hβ = 4 . ± . E ( B − V ) = 0 . ± .
01 (Izotov & Thuan 1998). c f Hα /f Hβ = 3 . ± . E ( B − V ) = 0 . ± .
01 (from our spectroscopic measurements). d f Hα /f Hβ = 4 . ± . E ( B − V ) = 0 . ± .
12 (from our spectroscopic measurements). e f Hα /f Hβ = 3 . ± . E ( B − V ) = 0 . ± .
01 (Izotov, Thuan & Lipovetsky 1994).References for 12 + log(O/H): f Izotov & Thuan (1998), g Guseva, Izotov & Thuan (2000), h Izotov & Thuan (1998), i Izotov, Thuan & Lipovetsky (1994), j Zhao et al. (2013), k Kunth & Joubert (1985), l Ekta & Chengalur (2010), m Izotov,Thuan & Lipovetsky (1994). The oxygen abundances for other galaxies are taken from Brinchmann et al. (2008).c (cid:13) , 1– ?? J a i s w a l &O m a r Table 3.
Summary of the optical observations.Galaxy name Telescope Date Exposure in H α Exposure in r FWHM PSF H α Sensitivity λ H α λ [min] [min] [arcsec] 10 − h erg s − cm arcsec i [˚A] [˚A]UM 311 DFOT 2012 Nov 09 150 10 2.0 1.47 6599.7 6570IC 225 DFOT 2012 Dec 08,09 220 30 2.3 0.88 6596.6 6570NGC 941 DFOT 2012 Dec 07,08 195 38 2.2 0.42 6598.2 6570NGC 1087 DFOT 2012 Nov 08,09 170 22 1.9 1.15 6596.2 6570UGCA 116 DFOT 2012 Nov 08,09 215 32 2.2 1.15 6580.3 6570UGCA 130 DFOT 2012 Dec 07 165 35 2.6 1.21 6580.3 6570NGC 2799 DFOT 2013 Feb 07 155 25 2.1 0.85 6599.6 6563MRK 22 DFOT 2012 Dec 08,09 235 50 2.1 0.92 6597.0 6570IC 2524 IGO 2012 Mar 19 117 15 1.3 1.20 6594.7 6563KUG 1013+381 IGO 2012 Mar 20 150 20 1.4 1.32 6588.7 6563NGC 3294 DFOT 2013 Feb 08 155 15 2.2 0.85 6597.7 6563NGC 3381 DFOT 2012 Apr 23 110 15 2.1 1.80 6598.7 6570NGC 3423 DFOT 2013 Mar 12 150 20 1.9 0.96 6585.1 6563NGC 3430 DFOT 2013 Mar 11 140 30 2.0 0.86 6597.7 6563CGCG 038-051 DFOT 2013 Mar 07 155 25 2.2 0.85 6585.4 6563IC 2828 IGO 2012 Mar 20 130 20 1.2 1.13 6585.7 6563UM 439 DFOT 2012 Apr 22,23 105 20 2.1 1.80 6587.1 6570NGC 3949 DFOT 2013 Mar 08 175 30 1.9 0.78 6580.5 6563I SZ 59 DFOT 2013 Mar 10 150 25 2.2 1.03 6609.7 6563CGCG 041-023 DFOT 2013 Feb 08 165 15 2.2 0.89 6592.6 6563SBS 1222+614 DFOT 2013 Mar 08,10 150 25 2.1 0.79 6578.5 6563NGC 4389 DFOT 2013 Mar 07,08 160 30 1.9 0.74 6578.7 6563IC 3521 DFOT 2012 Apr 22 89 15 2.5 1.90 6576.0 6570UGC 9273 IGO 2012 Mar 19 140 20 1.2 1.03 6591.2 6563MRK 475 DFOT 2012 May 20,21 170 23 2.2 0.98 6575.8 6570 c (cid:13) R A S , M N R A S , ?? α imaging survey of Wolf-Rayet galaxies Table 4.
Instrument calibration results.Telescope λ for H α filter FWHM for H α filter C Hα k Hα WNCR[˚A] [˚A] 10 − h erg s − cm − counts s − i [mag/airmass]DFOT 6570 77 31 . ± . . ± .
03 17 . ± . . ± . . ± .
02 13 . ± . . ± . . ± .
03 14 . ± . Table 5.
FUV, FIR and radio flux.Galaxy name F F UV S µ m S µ m S . [10 − erg s − cm − ˚A − ] [Jy] [Jy] [mJy]UM 311 1 . ± . . ± . A IC 225 1 . ± . < . A NGC 941 9 . ± . . ± . a . ± . a . ± . e NGC 1087 64 . ± . . ± . b . ± . b ∼ . f UGCA 116 3 . ± . . ± . a . ± . c . ± . e UGCA 130 1 . ± . . ± . a < . a < . g NGC 2799 4 . ± . < . d < . d ∼ . f MRK 22 0 . ± . . ± . A IC 2524 1 . ± . . ± . a . ± . a . ± . A KUG 1013+381 2 . ± . . ± . A NGC 3294 22 . ± . . ± . a . ± . a . ± . e NGC 3381 8 . ± . . ± . a . ± . a . ± . e NGC 3423 29 . ± . ∼ . f NGC 3430 30 . ± . . ± . a . ± . a . ± . e CGCG 038-051 0 . ± . . ± . A IC 2828 2 . ± . . ± . a . ± . a < . A UM 439 3 . ± . . ± . a < . a . ± . A NGC 3949 43 . ± . . ± . a . ± . a . ± . e I SZ 59 — 0 . ± . a . ± . a —CGCG 041-023 — — — 4 . ± . A SBS 1222+614 1 . ± . . ± . a < . a < . A NGC 4389 11 . ± . . ± . e IC 3521 2 . ± . . ± . a . ± . a . ± . A UGC 9273 1 . ± . < . A MRK 475 0 . ± . < . A References for S µ m and S µ m : a Moshir et al. (1990), b Soifer (1989), c Sanders et al. (2003), d Surace et al. (2004)References for S . : e Condon et al. (1998), f Condon et al. (2002), g Leroy et al. (2005). The values denoted bysuperscript ‘ A ’ are estimated using the FIRST images, while the others are taken from the NVSS catalogue.c (cid:13) , 1– ?? J a i s w a l &O m a r Table 6.
The H α flux. Galaxy name F Hα F iHα F iiHα F iiiHα [10 − erg s − cm − ] [10 − erg s − cm − ] [10 − erg s − cm − ] [10 − erg s − cm − ]UM 311 7 . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . ± ±
40 240 ±
41 274 ±
41 749 ± ±
11 188 ±
11 200 ±
11 1645 ± . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . ±
60 231 ±
60 262 ±
60 648 ± . ± . . ± . . ± . . ± . ±
51 306 ±
52 338 ±
52 454 ± ±
37 197 ±
37 224 ±
37 519 ± . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . ±
61 419 ±
62 464 ±
62 932 ± . ± . . ± . . ± . ± . ± . . ± . . ± . . ± . . ± . . ± . . ± . ± ±
28 130 ±
28 144 ±
28 293 ± . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . c (cid:13) R A S , M N R A S , ?? α i m a g i n g s u r ve y o f W o l f - R a y e t g a l a x i e s Table 7.
The luminosities and SFRs in different wave-bands.Galaxy name L F UV L Hα L F IR L . GHz
SF R
F UV
SF R Hα SF R
F IR
SF R . GHz
SF R assumed [10 erg s − ˚A − ] [10 erg s − ] [10 L ⊙ ] [10 W Hz − ] [ M ⊙ yr − ] [ M ⊙ yr − ] [ M ⊙ yr − ] [ M ⊙ yr − ] [ M ⊙ yr − ]UM 311 0 . ± . . ± . . ± . . ± .
01 0 . ± .
01 – 0 . ± .
002 0 . . ± . . ± . < . ± .
01 0 . ± .
03 – < . . ± . . ± . . ± . . ± . . ± .
09 0 . ± .
08 0 . ± .
03 0 . ± .
03 0 . . ± . . ± . . ± . . ± . . ± .
54 2 . ± .
60 1 . ± .
10 1 . ± .
08 2 . . ± . . ± . . ± . . ± . . ± .
01 1 . ± .
25 0 . ± .
02 0 . ± .
01 0 . . ± .
04 0 . ± . < < . ± .
002 0 . ± . < < . . ± . . ± . < . ± . . ± .
04 0 . ± . < . ± .
002 0 . . ± . . ± . . ± . . ± .
01 0 . ± .
01 – 0 . ± .
004 0 . . ± . . ± . . ± . . ± . . ± .
01 0 . ± .
02 0 . ± .
01 0 . ± .
002 0 . . ± . . ± . . ± . . ± .
01 0 . ± .
01 – 0 . ± .
002 0 . . ± . . ± . . ± . . ± . . ± .
21 2 . ± .
83 1 . ± .
10 0 . ± .
07 1 . . ± . . ± . . ± . . ± . . ± .
09 0 . ± .
10 0 . ± .
03 0 . ± .
01 0 . . ± . . ± . . ± . . ± .
11 0 . ± .
25 – 0 . ± .
01 0 . . ± . . ± . . ± . . ± . . ± .
28 2 . ± .
49 0 . ± .
06 0 . ± .
05 1 . . ± .
03 0 . ± . . ± . . ± .
003 0 . ± .
01 – 0 . ± .
01 0 . . ± . . ± . . ± . < . ± .
01 0 . ± .
02 0 . ± . < . . ± . . ± . < . ± . . ± .
01 0 . ± . < . ± .
004 0 . . ± . . ± . . ± . . ± . . ± .
10 1 . ± .
19 0 . ± .
02 0 . ± .
03 0 .
54I SZ 59 – 11 . ± . . ± . . ± .
18 0 . ± .
03 – 0 . . ± . . ± . . ± .
03 – 0 . ± .
004 0 . . ± .
03 1 . ± . < < . ± .
003 0 . ± . < < . . ± . . ± . . ± . . ± .
02 0 . ± .
06 – 0 . ± .
01 0 . . ± .
03 0 . ± . . ± . . ± .
04 0 . ± .
003 0 . ± .
01 0 . ± .
003 0 . ± .
001 0 . . ± . . ± . < . ± .
01 0 . ± .
01 – < . . ± .
01 0 . ± .
03 – < . ± .
001 0 . ± .
003 – < . c (cid:13) R A S , M N R A S , ?? Jaiswal & Omar
Table 8.
The q parameter and non-thermal to thermal radio flux ratio R.Name q log R UM 311 – 0 . ± . . ± .
30 0 . ± . . ± .
10 1 . ± . . ± . − . ± . ∼ .
65 –NGC 2799 < .
60 1 . ± . . ± . . ± .
31 0 . ± . . ± . . ± .
05 0 . ± . . ± .
11 1 . ± . . ± . . ± .
09 0 . ± . . ± . > .
09 –UM 439 < .
09 0 . ± . . ± .
04 0 . ± .
06I SZ 59 – –CGCG 041-023 – –SBS 1222+614 ∼ .
92 –NGC 4389 – 0 . ± . . ± .
11 0 . ± . (cid:13) , 1– ?? α i m a g i n g s u r ve y o f W o l f - R a y e t g a l a x i e s Table 9. H α and optical morphologies of galaxies.Name Type H α morphology optical morphology other features interaction probabilityUM 311 BCD lopsided SF interacting dwarf galaxy? lopsided radio continuum moderateIC 225 E nuclear un-disturbed — lowNGC 941 SAB(rs)c lopsided SF stellar bar, disturbed — moderateNGC 1087 SAB(rs)c lopsided SF misaligned H α and stellar bar — moderateUGCA 116 Irr pec arcs and plumes cometary, tidal tails merger highUGCA 130 Irr double nuclei cometary past interaction moderateNGC 2799 SB(s)m extra-planner H α regions lopsided, tidal tails close interacting pair highMRK 22 BCD arcs double nuclei merger moderateIC 2524 S nuclear — — lowKUG 1013+381 BCD nuclear — interacting pair moderateNGC 3294 SA(s)c cometary interacting dwarf galaxy? — lowNGC 3381 SB pec — bar, lopsided — lowNGC 3423 SA(s)cd — — disturbed radio lowNGC 3430 SAB(rs)c — — H i tidal features highCGCG 038-051 dIrr multiple nuclei? asymmetric — moderateIC 2828 Im two nuclei?, arcs and plumes cometary, irregular — moderateUM 439 Irr multiple nuclei?, arcs and plumes irregular H i lopsided moderateNGC 3949 SA(s)bc — bar offset radio emission moderateI SZ 59 S0 nuclear elongated envelope — lowCGCG 041-023 SB multiple nuclei? bar — lowSBS 1222+614 dIrr nuclear, arcs irregular — lowNGC 4389 SB(rs)bc pec extra-planner H α regions stellar bar, diffuse envelop — moderateIC 3521 IBm double nuclei, arcs bar, diffuse envelop — highUGC 9273 Im H α ring? lopsided — moderateMRK 475 BCD nuclear lopsided — low c (cid:13) R A S , M N R A S , ?? Jaiswal & Omar
Figure 1.
The histograms for size and metallicity of galaxies in our sample of WR galaxies. The vertical dashed line in themetallicity plot shows the solar metallicity, 12 + log(O/H) = 8.69. c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies Figure 2.
SDSS color images of galaxies in our sample. These images are made using g , r and i bands. Each image has a sizeof 3.4 ′ . North is up and East is to the left.c (cid:13) , 1– ?? Jaiswal & Omar T r an s m i tt an c e Wavelength [Å]
DFOT H a Filter: l = 6570 ÅDFOT H a Filter: l = 6563 ÅIGO H a Filter: l = 6563 Å Figure 3.
Transmittance vs. wavelength plot for different H α filters used in the present study.c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies
27 s30 s33 s 1 15 36 h m s- 0 52 ’ o 51 ’ 50 ’ 27 s30 s33 s 1 15 36 h m s- 0 52 ’ o 51 ’ 50 ’ 26 s27 s28 s29 s 2 26 30 h m s20" 40" +1 10 ’ 00" o 26 s27 s28 s29 s 2 26 30 h m s20" 40" +1 10 ’ 00" o 25 s27 s29 s 2 28 31 h m s- 1 09 ’ 50" o 20" 50" 20" 25 s27 s29 s 2 28 31 h m s- 1 09 ’ 50" o 20" 50" 20" 21 s24 s27 s 2 46 30 h m s- 0 31 ’ o 30 ’ 29 ’ 21 s24 s27 s 2 46 30 h m s- 0 31 ’ o 30 ’ 29 ’ 41 s43 s 5 55 45 h m s+3 22 ’ 50" o 20" 50" 41 s43 s 5 55 45 h m s+3 22 ’ 50" o 20" 50" 12 s16 s 6 42 20 h m s+75 37 ’ 25" o 40" 55" 12 s16 s 6 42 20 h m s+75 37 ’ 25" o 40" 55" 23 s28 s33 s 9 17 38 h m s58 ’ 59 ’ +42 00 ’ o 23 s28 s33 s 9 17 38 h m s58 ’ 59 ’ +42 00 ’ o 30 s31 s 9 49 32 h m s40" 50" +55 35 ’ 00" o 30 s31 s 9 49 32 h m s40" 50" +55 35 ’ 00" o 32 s 9 57 33 h m s+33 37 ’ 05" o 15" 25" 32 s 9 57 33 h m s+33 37 ’ 05" o 15" 25" 24 s10 16 25 h m s+37 54 ’ 35" o 40" 45" 50" 55" 24 s10 16 25 h m s+37 54 ’ 35" o 40" 45" 50" 55"
Figure 4. H α and r -band images of the WR galaxies in our sample. North is up and East is to the left. A linear scale-lengthin kpc is shown at the bottom of each image.c (cid:13) , 1– ?? Jaiswal & Omar
11 s16 s10 36 21 h m s+37 18 ’ o 19 ’ 20 ’ 11 s16 s10 36 21 h m s+37 18 ’ o 19 ’ 20 ’ 21 s24 s10 48 27 h m s+34 42 ’ 00" o 40" 20" 21 s24 s10 48 27 h m s+34 42 ’ 00" o 40" 20" 09 s14 s19 s10 51 24 h m s+5 48 ’ o 49 ’ 50 ’ 51 ’ 52 ’ 09 s14 s19 s10 51 24 h m s+5 48 ’ o 49 ’ 50 ’ 51 ’ 52 ’ 05 s10 s15 s10 52 20 h m s+32 56 ’ o 57 ’ 58 ’ 59 ’ 05 s10 s15 s10 52 20 h m s+32 56 ’ o 57 ’ 58 ’ 59 ’ 37 s38 s39 s40 s10 55 41 h m s20" 40" +2 24 ’ 00" o 37 s38 s39 s40 s10 55 41 h m s20" 40" +2 24 ’ 00" o 10 s11 s11 27 12 h m s+8 43 ’ 35" o 50" 05" 10 s11 s11 27 12 h m s+8 43 ’ 35" o 50" 05" 36 s37 s11 36 38 h m s+0 48 ’ 42" o 54" 06" 18" 36 s37 s11 36 38 h m s+0 48 ’ 42" o 54" 06" 18" 37 s42 s11 53 47 h m s+47 50 ’ 45" o 30" 15" 37 s42 s11 53 47 h m s+47 50 ’ 45" o 30" 15" 27 s28 s29 s11 57 30 h m s40" 20" - 19 37 ’ 00" o 27 s28 s29 s11 57 30 h m s40" 20" - 19 37 ’ 00" o 43 s45 s12 01 47 h m s+5 48 ’ 50" o 10" 30" 50" 43 s45 s12 01 47 h m s+5 48 ’ 50" o 10" 30" 50"
Figure 4. H α and r -band images of WR galaxies (continued).c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies
04 s05 s06 s12 25 07 h m s+61 09 ’ 00" o 10" 20" 04 s05 s06 s12 25 07 h m s+61 09 ’ 00" o 10" 20" 30 s34 s38 s12 25 42 h m s+45 40 ’ 10" o 50" 30" 10" 30 s34 s38 s12 25 42 h m s+45 40 ’ 10" o 50" 30" 10" 38 s40 s12 34 42 h m s+7 09 ’ 20" o 50" 20" 38 s40 s12 34 42 h m s+7 09 ’ 20" o 50" 20" 09 s10 s11 s12 s14 28 13 h m s+13 32 ’ 35" o 50" 05" 20" 35" 09 s10 s11 s12 s14 28 13 h m s+13 32 ’ 35" o 50" 05" 20" 35" 05 . 00 s 05 . 50 s 14 39 06 . 00 h m s +36 48 ’ 15" o 20" 25" 30" 05 . 00 s 05 . 50 s 14 39 06 . 00 h m s +36 48 ’ 15" o 20" 25" 30"
Figure 4. H α and r -band images of WR galaxies (continued).c (cid:13) , 1– ?? Jaiswal & Omar −3−2.5−2−1.5−1−0.5 0 0.5 1 1.5 −3 −2.5 −2 −1.5 −1 −0.5 0 0.5 1 1.5 l og ( S F R F U V ) [ l og ( M O · y r − ) ] log (SFR H a ) [log (M O· yr −1 )] (a) Correlation coefficient (r) = 0.90rms of residuals = 0.40
Linear fit: y = (1.11 ± 0.01)x + (−0.23 ± 0.01)SFR
FUV = SFR H a −3−2.5−2−1.5−1−0.5 0 0.5 1 1.5 −3 −2.5 −2 −1.5 −1 −0.5 0 0.5 1 1.5 l og ( S F R F I R ) [ l og ( M O · y r − ) ] log (SFR H a ) [log (M O· yr −1 )] (b) Correlation coefficient (r) = 0.88rms of residuals = 0.39
Linear fit: y = (1.29 ± 0.02)x + (−0.28 ± 0.01)SFR
FIR = SFR H a −3−2.5−2−1.5−1−0.5 0 0.5 1 1.5 −3 −2.5 −2 −1.5 −1 −0.5 0 0.5 1 1.5 l og ( S F R . G H z ) [ l og ( M O · y r − ) ] log (SFR H a ) [log (M O· yr −1 )] (c) Correlation coefficient (r) = 0.87rms of residuals = 0.52
Linear fit: y = (1.46 ± 0.01)x + (−0.39 ± 0.01)SFR = SFR H a Figure 5.
Comparison of H α -based SFRs with the SFRs derived using luminosities in other wavebands: (a) SF R
F UV vs.
SF R Hα (b) SF R
F IR vs.
SF R Hα (c) SF R . GHz vs.
SF R Hα . The dashed lines indicate equal SFRs in two wave-bands andthe solid lines are the error-weighted linear fits to the data. Small galaxies and large galaxies are designated by filled and opencircles, respectively. Other data points represented by filled square are from L´opez-S´anchez (2010). The upper-limits to theSFR values are denoted by arrows. c (cid:13) , 1– ?? α imaging survey of Wolf-Rayet galaxies l og R log (SFR FUV /SFR H a ) log R = 1.7log R = 0.9rms of residuals = 0.52 Linear fit: y = (0.47 ± 0.25)x + (0.95 ± 0.10)
Figure 6.
The plot of non-thermal to thermal radio flux ratio versus FUV to H α SFR ratio. The symbols have the samemeanings as in Figure 5. The solid line is the linear fit to the data. The dotted lines are the limits on log R for starburstgalaxies. −3−2.5−2−1.5−1−0.5 0 0.5 1 1.5 −3 −2.5 −2 −1.5 −1 −0.5 0 0.5 1 1.5 l og ( S F R . G H z ) [ l og ( M O · y r − ) ] log (SFR FIR ) [log (M O· yr −1 )] (a) Correlation coefficient (r) = 0.85rms of residuals = 0.30
Linear fit: y = (1.28 ± 0.01)x + (−0.10 ± 0.01)SFR = SFR
FIR q L FIR (10 L O • ) (b) q = 2.34q = 3.04q = 1.64 Figure 7.
The plots of radio based SFR versus FIR based SFR (left) and the radio-FIR correlation (right). In the left plot,symbols and lines have the same meanings as in Figure 5. In the right plot, top and bottom dotted lines are the limits of fivetimes FIR excess and five times radio excess from the mean q = 2 .
34 (solid line) respectively. Here, symbols have the samemeaning as in the left plot.c (cid:13) , 1–, 1–