A panchromatic spatially-resolved analysis of nearby galaxies -- I. Sub-kpc scale Main Sequence in grand-design spirals
A. Enia, G. Rodighiero, L. Morselli, V. Casasola, S. Bianchi, L. Rodriguez-Munoz, C. Mancini, A. Renzini, P. Popesso, P. Cassata, M. Negrello, A. Franceschini
MMNRAS , 1–14 (2020) Preprint 12 February 2020 Compiled using MNRAS L A TEX style file v3.0
A panchromatic spatially-resolved analysis of nearbygalaxies - I. Sub-kpc scale Main Sequence in grand-designspirals
A. Enia , (cid:63) , G. Rodighiero , , L. Morselli , , V. Casasola , , S. Bianchi ,L. Rodriguez-Mu˜noz , , C. Mancini , , A. Renzini , P. Popesso , P. Cassata , ,M. Negrello , A. Franceschini Dipartimento di Fisica e Astronomia, Universit`a di Padova, vicolo dell’Osservatorio 3, I-35122 Padova, Italy INAF − Osservatorio Astrofisico di Padova, vicolo dell’Osservatorio 5, I-35122 Padova, Italy INAF − Istituto di Radioastronomia, Via P. Gobetti 101, 40129, Bologna, Italy INAF − Osservatorio Astrofisico di Arcetri, Largo E. Fermi 5, 50125, Firenze, Italy Excellence Cluster Universe, Boltzmannstrasse 2, 85748, Garching bei Munchen, Germany School of Physics and Astronomy, Cardiff University, The Parade, Cardiff CF24 3AA, UK
Accepted XXX. Received YYY; in original form ZZZ
ABSTRACT
We analyse the spatially resolved relation between stellar mass (M (cid:63) ) and star for-mation rate (SFR) in disk galaxies (i.e. the Main Sequence, MS). The studied sampleincludes eight nearby face-on grand-design spirals, e.g. the descendant of high-redshift,rotationally-supported star-forming galaxies. We exploit photometric information over23 bands, from the UV to the far-IR, from the publicly available DustPedia databaseto build spatially resolved maps of stellar mass and star formation rates on sub-galacticscales of 0.5-1.5 kpc, by performing a spectral energy distribution fitting procedurethat accounts for both the observed and the obscured star formation processes, over awide range of internal galaxy environments (bulges, spiral arms, outskirts). With morethan 30 thousands physical cells, we have derived a definition of the local spatiallyresolved MS per unit area for disks, log ( Σ SFR ) =0.82log ( Σ ∗ ) -8.69. This is consistentwith the bulk of recent results based on optical IFU, using the H α line emission asa SFR tracer. Our work extends the analysis at lower sensitivities in both M (cid:63) andSFR surface densities, up to a factor ∼
10. The self consistency of the MS relationover different spatial scales, from sub-galactic to galactic, as well as with a rescaledcorrelation obtained for high redshift galaxies, clearly proves its universality.
Key words: galaxies: evolution – galaxies: spirals – galaxies: star formation
Galaxies appear to build their stellar masses in a steadymode mainly dominated by secular processes and thanks tothe accretion of cold gas. This picture finds its confirmationin the existence of a tight relation between the galaxy stellarmass (M (cid:63) ) and its star formation rate (SFR): the Main Se-quence (MS) of star-forming galaxies (SFGs), observed up toz ∼ ∼ (cid:63) E-mail: [email protected] gle et al. 2014; Whitaker et al. 2014; Shivaei et al. 2015;Schreiber et al. 2015; Renzini & Peng 2015; Santini et al.2017; Pearson et al. 2018; Popesso et al. 2019a). Galaxiesseem to oscillate around the MS relation as a consequenceof multiple events of central compaction of gas followed byinside-out gas depletion, thus related with the flows of coldgas in galaxies (Tacchella et al. 2015). Several works in therecent years exploited the MS relation as a reference to un-derstand the differences among galaxies characterised by dif-ferent rates of stellar production (starbursts, SFGs, passivegalaxies), with the final aim of understanding the originsof galaxy bimodality, and how the star formation activityis quenched (e.g. Rodighiero et al. 2011; Peng et al. 2015;Saintonge et al. 2016). © a r X i v : . [ a s t r o - ph . GA ] F e b Enia A. et al.
The existence of a tight relation between stellar masssurface density ( Σ (cid:63) ) and star formation rate surface den-sity ( Σ SFR ) found in HII regions of nearby galaxies, sug-gested that the global MS relation originates thanks to localprocesses that set the conversion of gas into stars (Rosales-Ortega et al. 2012; S´anchez et al. 2013). Such observationswere complemented by the work of Wuyts et al. (2013), thatdiscovered a correlation between Σ (cid:63) and Σ SFR on scales of1 kpc in galaxies at 0.7 < z < α flux as SFR tracer. Cano-D´ıaz et al.(2016) used the Calar Alto Legacy Integral Field Area Sur-vey (CALIFA, S´anchez et al. 2012) galaxies and found aspatially resolved MS on 0.5-1.5kpc scales with a slope of0.72 ± ∼ kpc scales. They find that the slope of the relation variesdramatically (from 0.3 to 0.99) depending on the range of Σ (cid:63) used for fitting, somehow recalling the bending of theMS observed in the integrated MS relation (e.g. Popessoet al. 2019a). In addition, they find a scatter that varies be-tween 0.55 and 0.7 and that they ascribe to a combinationof local variations of the specific SFR (sSFR) and differentlarge scale galaxy properties (i.e. morphology, existence ofa bar). Jafariyazani et al. (2019) found a spatially resolvedMS in galaxies at 0.1 < z < ∼
800 galaxiesin the Sydney AAO Multi-object Integral Field Galaxy Sur-vey (SAMI, Croom et al. 2012) at z < . , finding a slope of1.0. Hall et al. (2018) studied the spatially resolved MS in asample of 93 nearby galaxies drawn from the Survey for Ion-ized Neutral Gas in Galaxies (SINGG, Meurer et al. 2006)and the Wide-field Infrared Survey Explorer (WISE, Wrightet al. 2010) on scales ranging from 50pc to 10kpc. They findthat the slope, scatter and normalisation of the relation donot vary when changing the spatial scale, up to 1-1.5kpc.They also find no dependence of the scatter (of both thespatially resolved and the global relation) on structural pa-rameters of galaxies (morphology) and their HI gas fraction(as previously suggested for the global MS by, e.g., Saintongeet al. 2016). Thus, they conclude that the MS scatter can bedriven by systematic differences in the star formation pro-cess among galaxies or their global environment. Cano-D´ıazet al. (2019) exploited a sample of 2000 MaNGA galaxies tostudy the global and spatially resolved MS and its depen-dence on morphology. They find a spatially resolved MS witha slope close to 1 and a scatter of 0.27 dex, around whichSF areas are located depending on the galaxy integratedmorphology. They conclude that some processes related to morphology must set the local SF activity of a galaxy. Re-cently, Vulcani et al. (2019) studied the spatially resolvedMS in a sample of 30 local late-type star-forming galaxiesup to four effective radii, exploiting the GAs Stripping Phe-nomena in galaxies with MUSE survey (GASP, Poggiantiet al. 2017). They find that Σ (cid:63) and Σ SFR correlate with ascatter of 0.3 dex, larger than the one observed for CAL-IFA and MaNGA galaxies. Interestingly, this correlation isfound to vary dramatically when studied for single galaxies:for several sources a correlation is not even present, for oth-ers multisequences are observed, undermining the existenceof a universal correlation.To reconcile and homogenize all these observational re-sults into a coherent scenario, it is mandatory to accountfor the different selection samples and for potential system-atics in the methodology adopted to compute the galaxyphysical parameters. To this aim, we try to cope with themain biases affecting the bulk of the previous works that arealmost based on emission lines (i.e. H α ) or UV-to-opticaltracers to constrain the SFR (see Sanchez et al., 2019 for areview). Here, we exploit the power of a complete multiwave-length photometric coverage, extending from the far-UV upto the far-IR (i.e. from GALEX to Herschel ) to provide areliable measure of the comprehensive SFR in galaxies, di-rectly accounting for both the observed and the obscuredcomponents. This is obtained by performing an SED fittingprocedure, that, accounting for an energetic balance, allowsto go deeper than spectroscopic surveys based on H α emis-sion lines. The latters critically depend on dust extinctioncorrections derived from the Balmer decrement and mostlyon the spectral Signal-to-Noise ratio (SNR) limit needed (forexample) for the spaxels classification on BPT diagrams, re-quiring significant detection of four emission lines. Thus, weexploit the higher SNR of the photometric data that allowsto reach the farthest galactocentric distances, and the com-bination of UV and far-Infrared measurements to minimizethe uncertainties related to the presence of dust.With such a machinery, in this work (Paper I) we aimat defining the MS at kpc-scales in a sample of nearby face-on spirals collected in DustPedia (Davies et al. 2017). Thischoice is driven by the need to firstly characterize the fi-nal products of pure secular evolution processes, i.e. unper-turbed disks, that have assembled at earlier cosmic epochsthe bulk of the stars that we observe in today’s galax-ies (Rodighiero et al. 2011; Sargent et al. 2012). Indeed,grand-design spiral galaxies are considered to account forthe largest percentage of star forming galaxies sitting on thelocal MS (Wuyts et al. 2011, Morselli et al. 2017). This workfocuses first on the study of typical MS systems and it willbe extended in the future to other morphological type galax-ies, that are statistically located at larger distances from thelocal relation. To reach a comprehensive picture on the starformation process in galaxies, in a companion paper (PaperII, Morselli et al. in prep.) we will investigate possible varia-tions of the star formation efficiency (SFE) within galaxies,as well as the relation between cold gas availability and SFR.The paper is structured as follows. In Section 2 we de-scribe the dataset. In Section 3 we present our SED fittingmethodology (e.g. libraries, image processing). In Section 4we present our results on the spatially resolved MS. Thor-ough the paper we will assume a Λ CDM cosmology (PlanckCollaboration et al. 2016) and Chabrier (2003) IMF.
MNRAS , 1–14 (2020) . Sub-kpc MS in nearby spirals We exploit the vast amount of data from DustPedia , a col-lection of more than 875 local galaxies, built with the pur-pose of studying the dust emission in the local universe.DustPedia comprises every Herschel observed galaxy within ∼
50 Mpc from the Milky Way, with a diameter D > (cid:48) .The multiwavelength photometric data presented in DustPe-dia have been homogenised, making this dataset particularlyinteresting for our purposes. Starting from the DustPedia collection, we build a coherentsample of local star forming galaxies with a ”grand-designspiral” structure. Thus, we select galaxies having a Hubblestage index T (or RC3 type, de Vaucouleurs et al. 1991; Cor-win et al. 1994) between 2 and 8, and a diameter D > (cid:48) . Toavoid complications related to disk inclination ( i ) and dustcorrection, we focus here on nearly face-on sources with i < ◦ . Furthermore, we apply a cutoff in distance (around1000 km/s, corresponding to approximately 22 Mpc), as forgalaxies beyond this distance the sub-mm photometry isscarcely resolved or even below the observational beam. Fi-nally, to robustly estimate the physical parameters of galax-ies from SED fitting, we restrict our analysis to those havinga uniform coverage in the wavelength domain, with at least20 bands of observation.The final sample includes 8 objects: NGC0628,NGC3184, NGC3938, NGC4254, NGC4321, NGC4535,NGC5194, NGC5457. Distances, sizes, inclinations and mor-phologies are those adopted by the DustPedia collaboration,based on the HyperLEDA database (Makarov et al. 2014),and are available on their website (see Tab. 1). The samplestellar masses are in the range between − M (cid:12) , andstar formation rates between 1 and 4 M (cid:12) yr − These sourcesrepresent the evolved descendants of high-redshift normalstar-forming galaxies, that are rotationally supported sys-tems, at least by z = (e.g. Wisnioski et al. 2015; F¨orsterSchreiber et al. 2009, 2018). Moreover, their regular dynam-ical properties suggest that they have quiescently assembledtheir stellar mass through cosmic times. In Fig. 1 we present,as an example, the galaxy NGC0628 as it is observed inthe far UV (FUV) with the GALaxy Evolution eXplorer(GALEX, Morrissey et al. 2007), in the g-band with theSloan Digital Sky Survey (SDSS, York et al. 2000), in theJ-band with the 2 Micron All-Sky Survey (2MASS, Skrut-skie et al. 2006), at 8 µ m with the Spitzer
Space Telescope(Werner et al. 2004), at 12 µ m with WISE and finally at250 µ m with the Herschel
Space Observatory (Pilbratt et al.2010). In the following paragraphs we briefly report infor-mation on the different data sets used in this work. For amore systematic and complete reference about data reduc-tion and processing (implemented by the DustPedia collab-oration) we refer the reader to Clark et al. (2018). The DustPedia website is available athttp://dustpedia.astro.noa.gr The HyperLEDA database is available at http://leda.univ-lyon1.fr/
The ultraviolet part of the electromagnetic spectrum is sam-pled by GALEX. GALEX data are divided in near-UV(NUV, 1516 ˚A, with full-width at half maximun (FWHM)of ∼ . (cid:48)(cid:48) ) and FUV (2267 ˚A and ∼ . (cid:48)(cid:48) beam). NUV andFUV data are available for every galaxy in our sample, andsample the light coming from newborn massive stars, tracingthe unobscured star formation activity of galaxies. The optical part of the spectrum, sampling the young stellarcontent, is probed with the Data Release 12 of SDSS. We usefive bands ( u , g , r , i , z ), with effective wavelengths of 3351,4686, 6166, 7480 and 8932 ˚A. The SDSS FWHM varies fromobservation to observation, with a typical median value of1.43 (cid:48)(cid:48) . Near-infrared (NIR) imaging comes from 2MASS in threebands, J -band (1.25 µ m), H -band (1.65 µ m) and K s -band(2.16 µ m), with FWHMs between ∼ . (cid:48)(cid:48) and . (cid:48)(cid:48) . As notedin Appendix D of Clark et al. (2018), 2MASS-H band pho-tometry might be affected by a particular offset with respectto J and K s band photometry, translating as a ”bump” or”dip” in the SED of the object. We find this problem onlyNGC0628 photometry, for which we exclude the H-banddata-point from the SED fitting. The NIR and medium-infrared (MIR) parts of the spectrumare observed with the WISE and the IRAC camera on boardof
Spitzer . WISE provides observations at 3.4, 4.6, 12 and22 µ m, with Point Spread Function (PSF) FWHMs rangingfrom 5.70 (cid:48)(cid:48) to 11.8 (cid:48)(cid:48) . The IRAC camera, instead, collectsobservations at 3.56, 4.51, 5.76 and 8.00 µ m, with a PSFFWHM varying between 1.66 (cid:48)(cid:48) and 1.98 (cid:48)(cid:48) . These bands areavailable for the whole sample, except for NGC4535 andNGC5457, both lacking 5.8 µ m data. NIR and MIR observa-tions trace the old stellar component, the stellar mass dis-tribution and the carbonaceous-to-silicate materials in thedust. Emission from the far infrared (FIR) to the sub-mm is ob-served with the instruments on board the
Herschel
SpaceObservatory. We used both
Herschel
PACS (70 µ m, 100 µ mand 160 µ m) and SPIRE data (250 µ m and 350 µ m), withPSF FWHM of ∼
6, 8, 12, 18 and 24 (cid:48)(cid:48) respectively. The onlyexceptions are NGC4535 and NGC5194, lacking 70 µ m and100 µ m observations, respectively. These wavelengths typi-cally probe the reprocessed emission coming from dust, andthus could constrain the dust-obscured star-formation pro-cesses. We do not include the 500 µ m channel in our SED fit-ting procedure. Indeed, in our approach (see Sec. 3) we needto downgrade all images to the lower resolution available inour maps. With a beam of ∼ (cid:48)(cid:48) , the 500 µ m does not allowto push our spectro-photometric analysis to the kpc scale. MNRAS000
6, 8, 12, 18 and 24 (cid:48)(cid:48) respectively. The onlyexceptions are NGC4535 and NGC5194, lacking 70 µ m and100 µ m observations, respectively. These wavelengths typi-cally probe the reprocessed emission coming from dust, andthus could constrain the dust-obscured star-formation pro-cesses. We do not include the 500 µ m channel in our SED fit-ting procedure. Indeed, in our approach (see Sec. 3) we needto downgrade all images to the lower resolution available inour maps. With a beam of ∼ (cid:48)(cid:48) , the 500 µ m does not allowto push our spectro-photometric analysis to the kpc scale. MNRAS000 , 1–14 (2020)
Enia A. et al.
Figure 1.
Example of the multiwavelength imaging available for NGC0628, coming from the DustPedia archive. Image size is ∼ (cid:48) × (cid:48) . Table 1.
Our galaxy sample. Galaxy name, coordinates in J2000 system reference, distances D (in Mpc), inclinations, r sizes (thesemimajor axis isophote at which the optical surface brightness falls beneath 25 mag arcsec ) and morphological classifications are thesame adopted by the DustPedia collaboration, and come from the HyperLEDA database (Makarov et al. 2014). The values of M (cid:63) andSFR are obtained fitting the DustPedia photometry with magphys , the former as the standard magphys output, the latter from thescaling relations in Eq. 1-2. It is worth noticing that NGC5194 aperture photometry comprehends both fluxes coming from M51a andM51b. Reported cell sizes are referred to the so-called pixel-by-pixel SED fitting (in this case, with 8 (cid:48)(cid:48) square size, reported in kpc) andregions probing a fixed physical scale (1.5 kpc, in arcseconds), as reported in Sec. 3.Galaxy Name RA DEC D i r log M (cid:63) SFR Cell size RC3 Type[deg] [deg] [Mpc] [◦] kpc [ M (cid:12) ] [ M (cid:12) /yr] 8 (cid:48)(cid:48) [kpc] 1.5 kpc [ (cid:48)(cid:48) ]NGC 0628 (M74) 24.1740 15.7833 10.14 19.8 14.74 10.41 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± However, we have verified that the performed SED best-fitsare consistent with the observed 500 µ m flux densities. Theadopted spatial scale for our analysis will be discussed indetails in Sec. 3. To obtain the spatially resolved physical properties of galax-ies, like their stellar mass (M (cid:63) ) and SFRs, we perform a SEDfitting on scales varying from 0.28 kpc to 1.5 kpc.Our procedure includes three steps: (i) matching the PSF of every image to the worst-one, by degrading everyband to the PSF of the SPIRE 350 image; (ii) building agrid of square cells of a given size and measuring the flux ateach wavelength on them; (iii) deriving the physical proper-ties of the individual cells by performing SED fitting to theavailable photometry.We analyse each galaxy at two different resolutions: thefirst by considering cells of (cid:48)(cid:48) × (cid:48)(cid:48) (thus a varying lengthside in physical scale from one galaxy to another, as reportedin Tab.1), that is the pixel scale of SPIRE 350 maps, cor-responding to three spaxels per resolution element, and the MNRAS , 1–14 (2020) . Sub-kpc MS in nearby spirals second by constructing fixed size cells of 1.5 kpc × As reported in Sec. 2, the starting point are the different pho-tometric observations of the galaxies, downloaded from theDustPedia Archive. These have already been homogenised influx, given in Jy, and World Coordinate System. For our pur-poses, on each map we perform: foreground stars removal,background estimation, flattening and subtraction, and PSFdegradation, matching the maps resolution to the worst one.Foreground stars removal is performed exploiting theComprehensive & Adaptable Aperture Photometry Routine(CAAPR) routine presented in Clark et al. (2018). The rou-tine uses the PTS toolkit for SKIRT (Camps & Baes 2015)to detect and remove/patch foreground star emission, thuscreating a star subtracted version of the map. As CAAPRsometimes mistakenly identifies bright HII regions as stars,we carefully visual-check the star-subtracted maps to be surethat only foreground stars are removed.In each photometric band, we perform background es-timation and subtraction on the star-subtracted maps. Thisprocedure follows the indications in Clark et al. (2018), per-forming (if needed) a sky flattening fitting with a 5th orderpolynomial 2D array, and then subtracting the backgroundemission. This step is crucial to remove galactic foregroundemission and to smooth out residual image gradients typicalof some bands (i.e. GALEX, Spitzer).Finally, we degrade the stars and background sub-tracted maps to the PSF of SPIRE 350. To do this, we con-volve the maps using the kernels provided by Aniano et al.(2011) , in order to match each band to the PSF of SPIRE350. To perform SED fitting, we measure the flux in each pho-tometric band inside the cells with the photutils v0.6
Python package (Bradley et al. 2019). Then, we correct allthe fluxes with wavelength lower than 10 µ m for Galacticextinction, with the in-built module in CAAPR, based onvalues in the IRSA Galactic Dust Reddening and ExtinctionService. To estimate errors on the flux, we apply the follow-ing procedure. When available, we use the error maps in the ∼ ganiano/Kernels.html DustPedia Archive (i.e. Spitzer bands and the far-IR pho-tometry). If no error map is available (i.e. the SDSS maps),we take the signal-to-noise ratio of the DustPedia Photom-etry in that particular band (accounting also for calibrationerror), and use that to evaluate the error in each cell (thisis different from what done by Smith & Hayward (2018),that adopt a default SNR of 5 for every band). As a re-sult we obtain two photometric catalogues (for the (cid:48)(cid:48) × (cid:48)(cid:48) cell and for the 1.5kpc × magphys (da Cunha et al. 2008). magphys is one ofthe state-of-art code to model panchromatic SED, and assuch has been extensively used in the literature (e.g. daCunha et al. 2010; Smith et al. 2012; Berta et al. 2013; Hay-ward & Smith 2015; Smith & Hayward 2015; Chang et al.2015; Driver et al. 2018; Chrimes et al. 2018; Williams et al.2018; Martis et al. 2019). It simultaneously models the emis-sion observed in the UV-to-FIR regime by consistently as-suming that the whole energy output is balanced betweenthe one emitted at UV/optical/NIR wavelengths that is ab-sorbed by dust and the one re-emitted in the FIR. It does soexploiting a Bayesian approach to determine the posteriordistribution functions of the parameters by fitting the ob-served photometry to the model emission coming from a setof native libraries within the magphys distribution. Theselibraries are composed of 50000 stellar population spectrawith varying star forming histories (described by a continu-ous model ψ ( t ) ∝ exp − γ t with superimposed random bursts)derived from the Bruzual & Charlot (2003) spectral library.Formation ages are uniformly distributed between 0.1 and13.5 Gyr (thus they are shorter or equal to the age of theUniverse for the nearby galaxies in this work). The typicalSF timescales are uniform between 0 and 0.6 Gyr − , drop-ping exponentially around 1 Gyr − . Metallicities are takenfrom a uniform grid with values ranging between 0.02 and 2 Z (cid:12) . The stellar emission libraries are associated, via the en-ergy balance criterion, to 50000 two-components dust emis-sion SEDs (Charlot & Fall 2000; da Cunha et al. 2008),accounting for the emission coming from the so-called stel-lar birth clouds and the diffuse interstellar medium. mag-phys has already been used to model panchromatic SED oflocal galaxies (e.g. Viaene et al. 2014, for M31) and its abil-ity to properly retrieve physical properties from the SEDhas been tested on hydrodynamical simulations of galaxies(i.e. isolated disk or major merger as in Hayward & Smith2015; Smith & Hayward 2015) and of resolved regions (Smith& Hayward 2018). In particular, Smith & Hayward (2018)run magphys on 21 bands of a simulated isolated local diskgalaxy, on regions spanning from 0.2 to 10 kpc, also with dif-ferent disk inclinations, in order to test its ability to recoverphysical properties (such as M (cid:63) , SFRs, SFHs) and derivedones (i.e. metallicity, visual extinction A V ). They find that magphys is able to produce acceptable fits to almost everycell enclosed inside the r -band effective radius, and between59%-77% of cells within 20 kpc from the center (even though MNRAS000
Python package (Bradley et al. 2019). Then, we correct allthe fluxes with wavelength lower than 10 µ m for Galacticextinction, with the in-built module in CAAPR, based onvalues in the IRSA Galactic Dust Reddening and ExtinctionService. To estimate errors on the flux, we apply the follow-ing procedure. When available, we use the error maps in the ∼ ganiano/Kernels.html DustPedia Archive (i.e. Spitzer bands and the far-IR pho-tometry). If no error map is available (i.e. the SDSS maps),we take the signal-to-noise ratio of the DustPedia Photom-etry in that particular band (accounting also for calibrationerror), and use that to evaluate the error in each cell (thisis different from what done by Smith & Hayward (2018),that adopt a default SNR of 5 for every band). As a re-sult we obtain two photometric catalogues (for the (cid:48)(cid:48) × (cid:48)(cid:48) cell and for the 1.5kpc × magphys (da Cunha et al. 2008). magphys is one ofthe state-of-art code to model panchromatic SED, and assuch has been extensively used in the literature (e.g. daCunha et al. 2010; Smith et al. 2012; Berta et al. 2013; Hay-ward & Smith 2015; Smith & Hayward 2015; Chang et al.2015; Driver et al. 2018; Chrimes et al. 2018; Williams et al.2018; Martis et al. 2019). It simultaneously models the emis-sion observed in the UV-to-FIR regime by consistently as-suming that the whole energy output is balanced betweenthe one emitted at UV/optical/NIR wavelengths that is ab-sorbed by dust and the one re-emitted in the FIR. It does soexploiting a Bayesian approach to determine the posteriordistribution functions of the parameters by fitting the ob-served photometry to the model emission coming from a setof native libraries within the magphys distribution. Theselibraries are composed of 50000 stellar population spectrawith varying star forming histories (described by a continu-ous model ψ ( t ) ∝ exp − γ t with superimposed random bursts)derived from the Bruzual & Charlot (2003) spectral library.Formation ages are uniformly distributed between 0.1 and13.5 Gyr (thus they are shorter or equal to the age of theUniverse for the nearby galaxies in this work). The typicalSF timescales are uniform between 0 and 0.6 Gyr − , drop-ping exponentially around 1 Gyr − . Metallicities are takenfrom a uniform grid with values ranging between 0.02 and 2 Z (cid:12) . The stellar emission libraries are associated, via the en-ergy balance criterion, to 50000 two-components dust emis-sion SEDs (Charlot & Fall 2000; da Cunha et al. 2008),accounting for the emission coming from the so-called stel-lar birth clouds and the diffuse interstellar medium. mag-phys has already been used to model panchromatic SED oflocal galaxies (e.g. Viaene et al. 2014, for M31) and its abil-ity to properly retrieve physical properties from the SEDhas been tested on hydrodynamical simulations of galaxies(i.e. isolated disk or major merger as in Hayward & Smith2015; Smith & Hayward 2015) and of resolved regions (Smith& Hayward 2018). In particular, Smith & Hayward (2018)run magphys on 21 bands of a simulated isolated local diskgalaxy, on regions spanning from 0.2 to 10 kpc, also with dif-ferent disk inclinations, in order to test its ability to recoverphysical properties (such as M (cid:63) , SFRs, SFHs) and derivedones (i.e. metallicity, visual extinction A V ). They find that magphys is able to produce acceptable fits to almost everycell enclosed inside the r -band effective radius, and between59%-77% of cells within 20 kpc from the center (even though MNRAS000 , 1–14 (2020)
Enia A. et al. the SNR choice of 5 might play an important role in keep-ing the χ total value below the threshold). Those fits leadto reasonably good parameters estimate for every built-in magphys output.Following Smith & Hayward (2018), we input the mea-sured fluxes to magphys for SED fitting. As anticipated inSec. 3, we perform spatially resolved SED fitting for eachgalaxy using two different dimensions of the cell: 8 (cid:48)(cid:48) and1.5kpc. This is done for two reasons: i ) to compare the re-sults on different spatial scales, and ii ) to test the reliabil-ity of the pixel-by-pixel SED analysis. In fact, the physicalscales probed in this work vary from ∼ pc for NGC5457(the closest galaxy in our sample) to ∼ pc for NGC3938(the most distant, see Tab. 1). These scales are at the limitof the assumption on which the balance between star forma-tion laws and energy output still holds (e.g. Kravtsov 2003;Khoperskov & Vasiliev 2017; Schruba et al. 2010). In Sec. 4we show how the 8 (cid:48)(cid:48) results and the 1.5kpc ones are consis-tent, thus concluding that our analysis is reliable also on thesmallest scales.To accept or reject a fit, we apply a criterion basedon Hayward & Smith (2015), where they used a χ thresh-old of 20 on 17 simulated photometric bands. Differently, inSmith & Hayward (2018) the threshold is set to 30.6 on 21simulated bands. As the galaxies in our sample have beenobserved with 21 (NGC4535) or 23 photometric bands, weuse a conservative χ cut of 25. χ maps for each galaxyare reported in Fig. 2. Blue squares are the values belowthe threshold, while red points are the rejected ones. InNGC5194 the companion galaxy M51b has been removed byfurther rejecting each cell with a declination δ over 47.24052.On average, each galaxy has ∼ accepted points, withthe exceptions of the closest ones, NGC5194 ( ∼ ) andNGC5457 ( ∼ ). The output of the SED fitting process is a wide range ofphysical properties. To probe the MS, the two fundamen-tal quantities are the stellar mass and the star formationrate. M (cid:63) values are directly taken from the magphys out-put. SFRs are obtained by summing unobscured (SFR UV )and obscured (SFR IR ) contributions. SFR UV is estimatedusing the relation of Bell & Kennicutt (2001): SFR UV = . × − L ν , (1)where L ν in erg s − Hz − is evaluated from the SED fit at150 nm.SFR IR is computed using the relation of Kennicutt (1998): SFR IR = . × − L IR , (2)where L IR in erg s − is evaluated from the SED fit between 8and 1000 µ m (rest-frame). In both cases, the relations havebeen rescaled in accordance with a Chabrier (2003) IMF.We evaluate the SFRs from empirical relations that areless model dependent rather than using the SFRs obtainedfrom the SED fit that are more dependent on degenerateparameters like age, metallicity, extinction. A comparisonbetween the SFRs obtained from the empirical relations andthe ones given as output of magphys is shown in Fig.3. Thescatter of the relation is relatively small, but the SFRs from magphys are, on average, ∼ . dex smaller than the ones from the empirical relations. A detailed comparison of theoutput values of magphys with those from other tools orreceipts is beyond the goal of this paper. However, we notethat the SED fitting technique accounts for the contributionof different stellar populations at various ages, as determinedby the assumed SFHs. The empirical approach adopted here(and widely used in the literature) accounts for the contri-bution of two extreme populations: the UV emission fromyoung stars, and the totally obscured young stellar compo-nent still hidden in the dusty molecular clouds. The tightcorrelation found in Fig. 3 ensures that the use of the SFRfrom magphys would not change the results presented inthe next Section. There are some points falling below the1:1 relation, 369 with SFR increased by an order of magni-tude, and 134 by two orders of magnitude. This is due toattributed star forming activity to what the magphys mod-els identify as (strong) diffuse cirrus emission in the spiralarms and in the inter-arms regions (the spatial distribution isGaussian, centered on . r / r ). Anyway, we stress that theimpact of these points to the overall results is non-existent,being an extremely small fraction (1.72% and 0.62% respec-tively) of the full sample. Before deriving any general conclusions and implicationsfrom the statistical analysis of the combined sample of localface-on spirals introduced in this work, we briefly present themain results about each individual source. An example forNGC0628 is shown in Fig. 4 (the others are in Appendix A).In these Figures we present for each source: the maps of starformation rate, stellar mass and distance from the MS ( ∆ MS evaluated as the perpendicular distance of a point in thelog Σ (cid:63) -log Σ SFR from the MS relation), using as reference theMS evaluated for the combined sample), as well as the distri-bution of the cells in the log Σ (cid:63) - log Σ SFR plane, color coded asa function of the distance of the cell from the galaxy centre,in units of r . The dots in the maps correspond to rejectedcells (with χ larger than 25), while the color-coded squaresmark the cells with accepted χ (as explained in Sec. 3.2).The black solid line in the bottom right panel of Fig. 4 andFigures A1-A7 is the MS relation of the combined sample,as explained in Sec. 4.1.It is evident from the maps that the SFR traces the spi-ral pattern of each galaxy (as seen in the RGB image in theinset of the lower right panel). In addition, we can see thatfor almost all the sources the SFR distribution has a peakat the centre, as well as several other peaks along the spiralarms. The stellar mass distribution is smoother and morecentrally concentrated than as expected for an exponentialmass distribution; in fact, according to the morphologicalclassification of the galaxies as well as their RGB images, abulge component is present in all our sources. The map ofdistance from the MS relation reveals that the spiral armstend to lie above the relation for all the sources. The cen-tral bulge can be less (as in NGC0628, NGC3184, NGC3938,NGC5457) or more (as in NGC4254, NGC4321, NGC5194)starforming than the spiral arms. For those galaxies thathave a ’red’ bulge, a ’bending’ of the MS relation is ob-served at the high Σ (cid:63) end, as the cells corresponding to thebulge component are also the more massive ones. It is worth MNRAS , 1–14 (2020) . Sub-kpc MS in nearby spirals NGC0628 NGC3184 NGC3938 NGC4254NGC4321 NGC4535 NGC5194 NGC5457 χ Figure 2.
Final χ maps for the full sample of galaxies for cells of 8 (cid:48)(cid:48) . Blue colored squares are the accepted values (below the thresholdof 25, highlighted with a dotted line in the colorbar), red shaded points the discarded ones lying over the threshold. For NGC5194,each point with δ > . has been discarded in order to remove contamination from the M51b companion (they form a second tightsequence below the accepted cells). The dotted circle is the galaxy r reported in Tab. 1. Almost every cell in the innermost galaxyregions are accepted. In the outer part of the disk, the spiral arms tend to produce acceptable fits with respect to the non-spiral part ofthe galaxies, as it is clear from NGC5457. − − − − − − log SFR[M (cid:12) yr − ] − − − − − − l og S F R M A G P HY S [ M (cid:12) y r − ] Figure 3.
Comparison between the magphys output star-formation rates (y-axis) and the ones we used thorough this work,obtained from the empirical relations in Eqs. 1-2 (x-axis). The rederrorbar in the upper-left highlight the median error on the mea-sure. Black line is the 1:1 relation. The vast majority of the sampleis consistent with this relation, and only a small fraction (2.34%)see their SFR increase by more than one order of magnitude. noting that for the galaxies in our sample few spaxels wouldbe considered passive (1 dex below the MS). This is a conse-quence of the sample selection, made of grand-design spirals.Also, due to the sample selection, there are no morpholog- ical trends in the log Σ (cid:63) -log Σ SFR plane, i.e. in scatter fromthe global (or individual) MS or in slope. This aspect willbe the subject of a future work, where we plan to extendthe same analysis to all kind of morphological types withinDustPedia.The two galaxies that are infalling in the Virgo clus-ter, NGC4254 and NGC4321, show two peculiar character-istics. NGC4254 is almost entirely made of cells that are lo-cated above the spatially resolved MS relation; indeed, thisgalaxy is the one that, according to its integrated proper-ties, is located at the largest distance from the integratedMS relation (see Fig 5, X symbol), but still within the MSscatter. NGC4321 and NGC5194, instead, show the largestdifference in slope with respect to the average MS relation.Interestingly, it also shows a second ’quenched’ sequence be-low the MS, also observed in galaxies such as NGC4535 andNGC5194, and visible as a ring of red cells in Figures A4,A5 and A6.
In this Section, we analyse the spatially resolved MS us-ing two approaches: 1) by studying the distribution of thespatially resolved parameters in the logM (cid:63) - logSFR plane,and 2) by analysing the distribution in the log Σ (cid:63) - log Σ SFR plane. The first approach gives us the possibility to under-stand whether the integrated MS relations, computed usingSFR and M (cid:63) estimated over the full extent of galaxies, applyalso at lower scales, thus hinting towards the existence of auniversal relation at all scales. In Fig. 5 we show the logM (cid:63) - logSFR plane and how the different cell in which we de-
MNRAS000
MNRAS000 , 1–14 (2020)
Enia A. et al. ◦
15’ 12’ 09’ 06’15 ◦ RA [deg] D E C [ deg ] − . − . − . − . − . log SFR[ M (cid:12) /yr ] ◦
15’ 12’ 09’ 06’15 ◦ RA [deg] D E C [ deg ] − . − . − . − .
25 0 .
00 0 .
25 0 .
50 0 .
75 1 . Distance from MS ◦
15’ 12’ 09’ 06’15 ◦ RA [deg] D E C [ deg ] . . . . . . . log M ∗ [ M (cid:12) ] . . . . . . . . . log Σ ∗ [M (cid:12) kpc − ] − . − . − . − . − . − . − . − . − . − . l og Σ S F R [ M (cid:12) y r − k p c − ] R = . . . . . . . . Distance from center [1 /R ] NGC0628, Sc
Figure 4.
Summary plot for NGC0628. Panels are organized as follows: upper left , the cells log star formation rate; upper right , thecells distance from the MS defined in Sec. 4; lower left , the cells log stellar mass; lower right , how the galaxy cells are positioned in the Σ (cid:63) - Σ SFR plot, color-coded according to the distance from the galaxy center in units of r . Black line is the (total fitted) MS, red dashedline the ODR fit for the single galaxy points, orange lines the sensitivity limits, the inset an RGB image of the galaxy as observed inoptical bands. In the first three panels, cells rejected for having a χ over the threshold are showed as points. Results for the rest of thesample are reported in Appendix A. compose the galaxies in our sample populate this plane. Inparticular, we show the results obtained using two differentaperture sizes: 1.5 kpc (blue contours) and (cid:48)(cid:48) (red-to-yellowgradient). We also plot, with green symbols, the integratedlogM (cid:63) and logSFR values obtained via SED fitting on aper-tures that enclose the full galaxy. The orange dashed linesare the sensitivity thresholds, and correspond to the rms computed from the M ∗ and SFR maps on regions far awayfrom the galaxy emission. We have also checked that therms limit for the SFR is consistent with the one obtainedfrom the SPIRE maps. The red solid line is our linear fit ob-tained including only data points over the sensitivity thresh-olds (orange dashed lines); it has slope of 1.03 and intercept-10.17. The contours referring to different spatial scales and MNRAS , 1–14 (2020) . Sub-kpc MS in nearby spirals log M ∗ [M (cid:12) ] − − − − − l og S F R [ M (cid:12) y r − ] Whitaker+14
Integrated
NGC0628NGC3184NGC3938NGC4254 NGC4321NGC4535NGC5194NGC5457
Figure 5.
Sample properties evaluated in three different aper-tures. In green, the integrated properties, obtained fitting theavailable photometry in apertures containing the full galaxy emis-sion. Marker size is as big as the data error bars. Blue and filledred contours are referred to cells of 1.5 kpc and 8 (cid:48)(cid:48) respectively.Black line is the fit to the MS described by the 8 (cid:48)(cid:48) results. Pur-ple dashed line is the MS relation from Whitaker et al. (2014)rescaled at z ∼ , while the shaded area enclose the scatter of MSdistribution we obtain (0.27 dex). Dashed orange lines are the M (cid:63) and SFR sensitivity limit. the integrated quantities all lie along the the same relation.The logM (cid:63) - logSFR relation holds on a large variety ofscales, from ones slightly over the size of the Giant Molec-ular Clouds (GMCs, ∼
200 pc, Solomon et al. 1987) to thegalaxy as a whole. This results is corroborated by the recentwork of Chevance et al. (2019), showing that galactic starformation is driven by dynamical processes that are inde-pendent of the galaxy environment and are mostly governedby the GMC evolutionary cycling.In Fig. 5 we also show the MS relation of Whitaker et al.(2014, dashed purple line), that was originally computed us-ing integrated galaxy values in the redshift range [0.5:1.1]and for sources with stellar masses between 8.5 logM (cid:12) and11.5 logM (cid:12) . We also show the scatter of the MS relationderived for our sample (shaded region) at z ∼ , as moreevidence is mounting up for a non evolution of the MS scat-ter with redshift (e.g. Popesso et al. 2019b). We rescale theWhitaker MS using the redshift evolution given in Whitakeret al. (2014). Interestingly, despite the different proceduresand different stellar mass scales probed in the two works, theMS of Whitaker et al. (2014) is extremely consistent withthe one found in this paper. This agreement clearly indicatesthe existence of the integrated MS relation as a consequenceof a process that regulates the SF activity on small-scales.We then computed the MS relation in the log Σ (cid:63) -log Σ SFR plane considering all the accepted cells coming fromthe 8 galaxies. The fit, a log-linear relation log Σ SFR = m log Σ (cid:63) + q , is performed using emcee (Foreman-Mackey et al. 2013). Since we are dealing with a large number ofpoints (from ∼ for 1.5 kpc apertures to ∼ for 8 (cid:48)(cid:48) cells), we bin the data and perform linear fit on the binneddata points (gray points in Fig. 6). Whenever binning, wediscard the points falling below the sensitivity limits andconsider only the ones above. The mass bins are built in twosteps: first we subdivide the log Σ (cid:63) interval in 20 bins, eachcontaining the same number of points. In order to probe theupper part of the MS (that otherwise would not have anybinned points) we add a bin between . ≤ Σ (cid:63) < and oneover 9 M (cid:12) yr − kpc − . Within each bin, we compute themedian Σ (cid:63) and Σ SFR , while the error is taken as the stan-dard deviation of the points inside the bin. We fit the binneddata points with a log -linear relation. In Fig. 6 we report the8 (cid:48)(cid:48) cells results for the sample, and the fit to the data (leftpanel). The results of the fit are a slope of . ± . andintercept of − . ± . . As expected from Fig. 3, we testedthat the MS obtained fitting the SFRs coming directly from MAGPHYS is consistent (almost identical) with this one.Our slope is consistent with others in literature (seeSec. 4.2) that usually span the range . − . despite us-ing different SFR tracers and different redshift and stellarmass ranges for the analysis (i.e. P´erez et al. 2013; Wuytset al. 2013; Hemmati et al. 2014; Cano-D´ıaz et al. 2016;Gonz´alez Delgado et al. 2016; Abdurro’uf & Akiyama 2017;Abdurro’uf 2018; Hall et al. 2018; Cano-D´ıaz et al. 2019). Itis worth noting that several of the literature works imple-ment the orthogonal distance regression (ODR) method tofind the location of the MS relation. If we perform an ODRfit on our results, we obtain a slope of 0.89 and an interceptof -9.13.The right panels of Fig. 6 are dedicated to the distribu-tion scatter σ . We investigate if the scatter of the spatiallyresolved MS varies as a function of mass by dividing oursample in three mass bins, . ≤ log Σ (cid:63) [ M (cid:12) yr − ] < . (redhistogram), . ≤ log Σ (cid:63) [ M (cid:12) yr − ] < . (green histogram), log Σ (cid:63) > . (cid:12) (blue histogram). The total scatter of thefull sample (colored histogram) is σ = . . There are hintsof a decreasing scatter with increasing stellar mass, from0.35 to 0.23. Once again, these values fall within the typicalranges of 0.15-0.35 reported in the literature for the spa-tially resolved MS (Conselice et al. 2016; Magdis et al. 2016;Maragkoudakis et al. 2017; Hall et al. 2018). The importanceof the scatter and the way it relates with the gas propertiesis explored in Morselli et al. (in prep).Finally, we check that the results are consistent on dif-ferent spatial scales. In Fig. 7 we compare the surface densi-ties obtained for 8 (cid:48)(cid:48) cells (scales varying from 0.2 to 0.8 kpc)and 1.5 kpc cells. The black solid line is the MS fit to 8 (cid:48)(cid:48) data(though the fit to 1.5 kpc data lead to the same values wellwithin the uncertainties), orange lines the sensitivity thresh-olds. Both distributions follow the same relation, and occupythe same region of the plane, being clear how the retrievedsurface density properties on bigger scales reflects the aver-age surface density properties in enclosed smaller scales. Themain differences arise for galaxy regions where χ thresholdis not reached, as expected since 1.5 kpc have an intrinsi-cally higher signal-to-noise ratio for the SED points, alwaysin regions below the sensitivity thresholds. The fact that thetwo realizations in different physical scales (sometimes evenof an order of magnitude, i.e. NGC5457) are almost over- MNRAS , 1–14 (2020) Enia A. et al. σ = . < log Σ ∗ < . σ = . < log Σ ∗ < . − . − . . . . Distance from MS σ = log Σ ∗ > . − . − . . . . Distance from MS σ = log Σ ∗ [M (cid:12) kpc − ] − − − − − l og Σ S F R [ M (cid:12) y r − k p c − ] − . − . − . − . . . . . . D i s t a n ce f r o m M S Figure 6.
The fitted main sequence, and its scatter.
Left panel: (cid:48)(cid:48) cells results for star formation rate surface density and stellar massdensity, color coded as a function of their distance from the main sequence. Orange dashes lines are the sensitivity limits. The MS (blacksolid line) is obtained as the linear fit of the gray data points. The error bars are the 1 σ dispersion in each bin. Right panel: distributionof distances from the MS, in three stellar mass bins (red, cyan and blue, color coded as in the left panel). The σ values reported in eachpanel are obtained by fitting the distribution with a Gaussian. The scatter varies between 0.22 for the highest mass bin to 0.35 for thelowest mass bin. log Σ ∗ [M (cid:12) kpc − ] − − − − l og Σ S F R [ M (cid:12) y r − k p c − ] Pixel-by-pixel1.5 kpc
Figure 7. Σ ∗ - Σ SFR results for 8 (cid:48)(cid:48) (red points) and 1.5 kpc (bluecircles) cells. Dashed lines are the sensitivity thresholds. It is clearhow the results coming from the higher 1.5 kpc physical scale arethe mean of the properties coming from the single cells insidethem. Both data sets lead to the same linear relations (red lineand blue dashed line respectively), well within the uncertainties. lapping tells us that the star formation laws still hold onphysical scales lower than ∼ pc. In recent years, as already mentioned in the introduction,several works have analysed the spatially resolved MS ofSFGs, thanks to the increased availability of integral fieldspectroscopic data. Surveys like MaNGA, CALIFA andSAMI made it possible to obtain the SFR from the H α lu-minosity, and to correct it for dust absorption through theBalmer decrement. The lack of information on the IR emis-sion, however, might prevent a comprehensive evaluation ofthe energetic budget in galaxies, as UV and optical tracerscould underestimate a fraction of the obscured SFR (e.g.Rodighiero et al. 2014). In this Section, we thus compareour panchromatic results with the spatially resolved MS re-lations obtained with data from CALIFA (Cano-D´ıaz et al.2016), MaNGA (Hsieh et al. 2017; Cano-D´ıaz et al. 2019;Bluck et al. 2019), and SAMI (Medling et al. 2018). For com-pleteness, we also compare our results with the MS relationsof Abdurro’uf & Akiyama (2017), that perform pixel-by-pixel SED fitting to GALEX and SDSS photometry of local( . < z < . ) massive spiral galaxies selected in the MPI-JHU (Max Planck Institute for Astrophysics-Johns HopkinsUniversity), and of Hall et al. (2018), obtained for a sampleof 355 nearby galaxies, with spatially resolved observationsof H α and mid-IR emission.In Fig. 8 we show the spatially resolved MS relation ofthis work (solid black line) together with the relations men-tioned above. To underline the depth of each study, in Fig. 8we plot the relations as starting from the logM (cid:63) value abovewhich 80 % of the corresponding data are located. The re-lations have been homogenised in terms of IMF and cos-mology, and rescaled to the median redshift of our sample, MNRAS , 1–14 (2020) . Sub-kpc MS in nearby spirals Table 2.
Summary of the slopes quoted in the text, with associ-ated SFR tracer.Reference Slope SFR tracerCano-D´ıaz et al. (2016) 0.72 H α Abdurro’uf & Akiyama (2017) 0.99 SED fittingHsieh et al. (2017) 1.00 H α Maragkoudakis et al. (2017) 0.91 3.6 µ m or 8.0 µ mHall et al. (2018) 0.99 H α + 24 µ mMedling et al. (2018) 1.00 H α Cano-D´ıaz et al. (2019) 0.94 H α Bluck et al. (2019) 0.90 H α This Work 0.82 SED fitting using the evolution in MS normalisation of SFR ∝ (1+z) . (Whitaker et al. 2014).The spatially resolved MS of Cano-D´ıaz et al. (2016,lavender dashed line) has been estimated from a sample of306 galaxies with mixed morphologies at . < z < . , onspatial scales of 0.5-1.5 kpc. They find a slope of 0.72. Sim-ilarly, Hsieh et al. (2017) used 536 SFGs at . < z < . from the MaNGA survey to obtain a MS relation on scalesof ∼ ∼ . < z < . )to probe the spatially resolved MS for a larger sample, againon spatial scales of ∼ ∼ lo-cal galaxies in the SDSS-IV MaNGA-DR15 data, with SFRscoming from D4000 and H α observations whenever available,Bluck et al. (2019, yellow line) obtain a slope of 0.90 on adataset of over 5 million spaxels. Maragkoudakis et al. (2017,blue dotted line) evaluate SFRs from IRAC 3.6 and 8.0 µ mdata at sub-kpc scales on a sample of 369 nearby galaxies( z ∼ . ), obtaining a slope of 0.91. The MS of Medlinget al. (2018, purple solid line) has been estimated from ∼
800 galaxies in the SAMI Galaxy Survey at z < . and hasa slope of 1.0. The MS relation of Hall et al. (2018, magentadashed line) has a slope of 0.99, and was computed exploitingspatially resolved H α observations of nearby galaxies withinthe Survey of Ionization in Neutral Gas Galaxies (SINGG)and WISE surveys (we report here the relation obtained us-ing the SFR transformation of Calzetti et al. 2007). Finally,the MS of Abdurro’uf & Akiyama (2017, red line) has a slopeof 0.99.The immediate take-home information evident fromFig. 8 is that thanks to our multiwavelength approach weare able to probe regions of lower stellar mass surface densi-ties (up to a factor of 10) compared to spectroscopic obser-vations, bound to the sensitivity limit of the H α line. Thistranslates in the ability of sampling regions that are locatedfurther away from the galaxy centre, up to ∼ α line with a signal-to-noise ratio (S/N) > . . . . . . . . log Σ ∗ [M (cid:12) kpc − ] − . − . − . − . − . − . − . l og Σ S F R [ M (cid:12) y r − k p c − ] This WorkAbdurro’uf & Akiyama 17Hsieh+17Maragkoudakis+17Medling+18Hall+18 (Calzetti SFR)Cano-Diaz+16Cano-Diaz+19Bluck+19Jafariyazani+19
Figure 8.
A comparison between our results (black-to-yellow con-tours encircling from 80% to 10% of our data points, fitted by thesolid black line) and a sample of resolved MS relations from theliterature, respectively. Each relation, that has been converted toa Chabrier (2003) IMF, and re-scaled to z = 0 (as described inthe text), starts from the sensitivity limit reported in the study(or, if not, estimated from the relative data). (2019). On the other hand, Hall et al. (2018) reach lower Σ SFR thanks to the combination of H α and mid-IR data, buthas the drawback that low Σ SFR only trace obscured SF.The MS slope that we obtain is lower than the onesshown in Fig. 8 with the exception of the Cano-D´ıaz et al.(2016) relation. To exclude effects due to different fittingprocedures, that can largely affect the MS slope and inter-cept, we also compute the relation using the ODR method,and we find a slope of 0.88. This value is still lower thanthe ones found in the works mentioned above, but it is im-portant to underline that our relation was computed from8 galaxies against the hundreds of sources of other works.As there are strong galaxy-to-galaxy variations, it is impor-tant to apply our approach on a larger sample to carry outa more meaningful comparison with other works.
In this paper we have presented a new analysis of eight lo-cal face-on spirals, with the aim of understanding if the MSis a universal relation holding also on sub-galactic scales.Other surveys of nearby galaxies have addressed this ques-tion, showing clear correlation between stellar masses andstar-formation per unit area, and to their gas content (Bigielet al. 2008; Leroy et al. 2008; Casasola et al. 2015). By ex-ploiting the publicly available photometric information ofDustPedia in the UV to far-Infrared spectral range, we havebeen able to perform a global SED fitting procedure that al-lowed us to simultaneously account for a careful evaluationof the obscured and unobscured SFR components, over thefull optical radius, larger than what obtained with optical
MNRAS000
MNRAS000 , 1–14 (2020) Enia A. et al.
IFS at similar redshifts. Our limited sample is restricted tothe grand design spirals with low inclination, large spatialextension and regular spiral arms structures in DustPedia.This analysis has provided a total of tens of thousands phys-ical cells on typical scales of ∼ . kpc over very different in-ternal galaxy environments (bulges, spiral arms, inter-armsregions, outskirts). This set is thus well fit to study the sec-ular evolution processes that regulate the star formation inlocal sources, dominated by rotationally supported systems(e.g. Law et al. 2009; F¨orster Schreiber et al. 2009; Glaze-brook 2013; Wisnioski et al. 2015; Simons et al. 2017; F¨orsterSchreiber et al. 2018; ¨Ubler et al. 2019).Some individual galaxies show peculiar variation aroundthe Σ SFR − Σ (cid:63) relation presented in Fig. 6 (see bottom-right panels in Figs A1-A8). For example, NGC3938 andNGC4254 show an average enhancement of the SFR, that weinterpret as a possible effect of the environment, as these twosources are located in the Ursa Major group and the Virgocluster, respectively. Other galaxies, in particular NGC0628,show a prominent bulge feature appearing as a narrow dis-tribution at the highest stellar mass densities. Indeed, wealready mentioned that at the lowest stellar mass densities(i.e. at large galactocentric distances) in few sources we haveidentified a cloud of cells deviating from the main relation to-ward lower SFR, at fixed stellar mass (NGC4321 is the clean-est example). Such distributions are circularly distributed inthe outer parts of the galaxies, out of the regions spannedby the dynamical interaction of the spiral arms. It is stillto be understood if this is associated to an older populationmigrated out of the disk, or if it is the remnant of exter-nal accretion through, for example, minor merging. Finally,only NGC5457 reveals strong mini-starbursts inside the spi-ral arms (i.e. regions well elevated above the MS, by a factorlarger then 10-100 times at fixed mass density).In any case, the combination of all the eight galaxiesdemonstrates the existence of a universal relation, as the de-viation of single sources is well within the global scatter of ∼ . dex (see Sect. 4.3). We have then demonstrated thatsuch relation holds at different galaxy scales, supporting theinterpretation from other surveys that the SFR is regulatedby local processes of gas-to-stars conversion happening atGMC scales on the order of a few hundred pc. With respectto other works, we have however provided evidence that such secular regulation keeps to apply at the farthest galactocen-tric distances, where the optical disk is still influenced bythe density waves motions, originating the spiral arms.The present work defines an accurate locus in the Σ SFR − Σ (cid:63) plane, constrained by observed distributions cov-ering 3 dex in the parameters space. Such a definition of thespatially resolved MS in local disk galaxies represents a valu-able reference for future comparison to different galaxy mor-phological types adopting a similar panchromatic approach.Indeed, there are indication that galaxy morphology playsan important role in the characterising the spatially resolvedMS relation, especially its scatter (e.g. Gonz´alez Delgadoet al. 2016; Cano-D´ıaz et al. 2019). Maragkoudakis et al.(2017) observe a decrease in the spatially resolved MS re-lation from late-type to early-type spirals, while the scatterremains constant. Other works do not find a similar connec-tion (Hall et al. 2018). These examples show the importanceof extending the analysis performed in this work to a largersample of galaxies encompassing different morphologies. In the second paper of this series, we will exploit thisreference sample to study if and how the distance to theresolved MS is connected to the total (atomic and molecular)gas content (Morselli et al., in prep.). The star-forming main sequence is a well studied tight re-lation between stellar masses and star formation rates, ob-served up to z ∼ , over a great variety of environments andmorphologies, both globally, counting galaxies as a whole,and locally, resolving physical properties in single galaxy re-gions. In this work we perform spatially resolved SED fittingin a sample of 8 local grand-design spirals taken from theDustPedia archive and we use the outputted maps of stellarmass and star formation rate (see Appendix) to analyse thespatially resolved MS of SFGs on scales spanning the rangebetween 0.4-1.5 kpc. We summarise here our main findings: • When considering the 8 galaxies together, we obtain aspatially resolved Main Sequence with a slope of 0.82 andan intercept of -8.69. When fitting the data with the ODRmethod we obtain a slope and an intercept of 0.88 and -9.05,respectively. This relation holds on different scales, from sub-galactic to galactic; • The local spatially resolved MS is consistent with theevolutionary (from high-z) low-mass relation, thus provingits universality across cosmic time. This is a crucial point tovalidate the integrated information on individual galaxies athigh redshift; • We have overtaken the limits of all the spectroscopicresolved emission lines studies, based mainly on H α andBalmer decrement corrections for dust extinction. The sen-sitivities of these surveys do not sample the lowest M (cid:63) andSFR as we do with a multiwavelength photometric approach,allowing us to probe the outermost regions of galaxies.We plan to extend this analysis on different morpholog-ical types inside the DustPedia sample, while improving theSED fitting pipeline to investigate star formation historiesof resolved galaxy regions. We will also link these resultswith existing gas observations (i.e. CO, H , CII) to furtherunderstand the role of secular evolution in galaxies. ACKNOWLEDGEMENTS
We would like to thank the anonymous referee for the use-ful comments improving the overall work quality. We aregrateful to CJR Clark for the information on CAAPR,and to M. Cano-Diaz for providing us the CALIFA resultsfrom her 2016 paper. AE and GR are supported from theSTARS@UniPD grant. GR acknowledges the support fromgrant PRIN MIUR 2017 - 20173ML3WW 001. GR and CMacknowledge funding from the INAF PRIN-SKA 2017 pro-gram 1.05.01.88.04. We acknowledge funding from the INAFmain stream 2018 program ”Gas-DustPedia: A definitiveview of the ISM in the Local Universe”. This research isbased on observations made with the Galaxy Evolution Ex-plorer, obtained from the MAST data archive at the SpaceTelescope Science Institute, which is operated by the Associ-ation of Universities for Research in Astronomy, Inc., under
MNRAS , 1–14 (2020) . Sub-kpc MS in nearby spirals NASA contract NASA 5-26555. DustPedia is a collaborativefocused research project supported by the European Unionunder the Seventh Framework Programme (2007-2013) call(proposal no. 606847). The participating institutions are:Cardiff University, UK; National Observatory of Athens,Greece; Ghent University, Belgium; Universit´e Paris Sud,France; National Institute for Astrophysics, Italy and CEA(Paris), France. This research made use of Photutils, an As-tropy package for detection and photometry of astronomicalsources.
REFERENCES
Abdurro’uf Akiyama M., 2018, MNRAS, 479, 5083Abdurro’uf Akiyama M., 2017, MNRAS, 469, 2806Aniano G., Draine B. T., Gordon K. D., Sandstrom K., 2011,PASP, 123, 1218Bell E. F., Kennicutt Robert C. J., 2001, ApJ, 548, 681Berta S., et al., 2013, A&A, 551, A100Bigiel F., Leroy A., Walter F., Brinks E., de Blok W. J. G.,Madore B., Thornley M. D., 2008, AJ, 136, 2846Bluck A. F. L., Maiolino R., S´anchez S. F., Ellison S. L., ThorpM. D., Piotrowska J. M., Teimoorinia H., Bundy K. A., 2019,MNRAS, p. 2839Bradley L., et al., 2019, astropy/photutils: v0.6,doi:10.5281/zenodo.2533376, https://doi.org/10.5281/zenodo.2533376
Brammer G., et al., 2012, arXiv.org, p. 13Bruzual G., Charlot S., 2003, Monthly Notices of the Royal As-tronomical Society, 344, 1000Bundy K., et al., 2015, The Astrophysical Journal, 798, 7Calzetti D., et al., 2007, ApJ, 666, 870Camps P., Baes M., 2015, Astronomy and Computing, 9, 20Cano-D´ıaz M., et al., 2016, The Astrophysical Journal Letters,821, L26Cano-D´ıaz M., ´Avila-Reese V., S´anchez S. F., Hern´andez-ToledoH. M., Rodr´ıguez-Puebla A., Boquien M., Ibarra-Medel H.,2019, MNRAS, 488, 3929Casasola V., Hunt L., Combes F., Garc´ıa-Burillo S., 2015, A&A,577, A135Chabrier G., 2003, The Publications of the Astronomical Societyof the Pacific, 115, 763Chang Y.-Y., van der Wel A., da Cunha E., Rix H.-W., 2015,ApJS, 219, 8Charlot S., Fall S. M., 2000, ApJ, 539, 718Chevance M., et al., 2019, arXiv e-prints, p. arXiv:1911.03479Chrimes A. A., Stanway E. R., Levan A. J., Davies L. J. M.,Angus C. R., Greis S. M. L., 2018, MNRAS, 478, 2Clark C. J. R., et al., 2018, A&A, 609, A37Conselice C. J., Wilkinson A., Duncan K., Mortlock A., 2016, TheAstrophysical Journal, 830, 83Corwin Harold G. J., Buta R. J., de Vaucouleurs G., 1994, AJ,108, 2128Croom S. M., et al., 2012, Monthly Notices of the Royal Astro-nomical Society, 421, 872Daddi E., et al., 2007, The Astrophysical Journal, 670, 156Davies J. I., et al., 2017, PASP, 129, 044102Driver S. P., et al., 2018, MNRAS, 475, 2891Elbaz D., et al., 2007, Astronomy & Astrophysics, 468, 33Foreman-Mackey D., Hogg D. W., Lang D., Goodman J., 2013,PASP, 125, 306F¨orster Schreiber N. M., et al., 2009, ApJ, 706, 1364F¨orster Schreiber N. M., et al., 2018, ApJS, 238, 21Glazebrook K., 2013, Publ. Astron. Soc. Australia, 30, e056Gonz´alez Delgado R. M., et al., 2016, A&A, 590, A44Grogin N. A., et al., 2011, ApJS, 197, 35 Hall C., Courteau S., Jarrett T., Cluver M., Meurer G., CarignanC., Audcent-Ross F., 2018, ApJ, 865, 154Hayward C. C., Smith D. J. B., 2015, MNRAS, 446, 1512Hemmati S., et al., 2014, ApJ, 797, 108Hsieh B. C., et al., 2017, ApJ, 851, L24Jafariyazani M., Mobasher B., Hemmati S., Fetherolf T., Khos-tovan A. A., Chartab N., 2019, ApJ, 887, 204Kennicutt Robert C. J., 1998, ApJ, 498, 541Khoperskov S. A., Vasiliev E. O., 2017, MNRAS, 468, 920Koekemoer A. M., et al., 2011, The Astrophysical Journal Sup-plement, 197, 36Kravtsov A. V., 2003, ApJ, 590, L1Law D. R., Steidel C. C., Erb D. K., Larkin J. E., Pettini M.,Shapley A. E., Wright S. A., 2009, ApJ, 697, 2057Leroy A. K., Walter F., Brinks E., Bigiel F., de Blok W. J. G.,Madore B., Thornley M. D., 2008, AJ, 136, 2782Lin L., et al., 2017, ApJ, 851, 18Magdis G. E., et al., 2016, Monthly Notices of the Royal Astro-nomical Society, 456, 4533Makarov D., Prugniel P., Terekhova N., Courtois H., Vauglin I.,2014, A&A, 570, A13Maragkoudakis A., Zezas A., Ashby M. L. N., Willner S. P., 2017,Monthly Notices of the Royal Astronomical Society, 466, 1192Martis N. S., Marchesini D. M., Muzzin A., Stefanon M., Bram-mer G., da Cunha E., Sajina A., Labbe I., 2019, ApJ, 882,65Medling A. M., et al., 2018, Monthly Notices of the Royal Astro-nomical Society, 475, 5194Meurer G. R., et al., 2006, ApJS, 165, 307Morrissey P., et al., 2007, ApJS, 173, 682Noeske K. G., et al., 2007, The Astrophysical Journal, 660, L47Oliver S., et al., 2010, MNRAS, 405, 2279Pannella M., et al., 2009, arXiv.org, pp L116–L120Pearson W. J., et al., 2018, A&A, 615, A146Peng Y., Maiolino R., Cochrane R., 2015, Nature, 521, 192P´erez E., et al., 2013, ApJ, 764, L1Pilbratt G. L., et al., 2010, A&A, 518, L1Planck Collaboration et al., 2016, A&A, 594, A13Poggianti B. M., et al., 2017, ApJ, 844, 48Popesso P., et al., 2019a, MNRAS, 483, 3213Popesso P., et al., 2019b, MNRAS, 490, 5285Renzini A., Peng Y.-j., 2015, The Astrophysical Journal Letters,801, L29Rodighiero G., et al., 2011, The Astrophysical Journal Letters,739, L40Rodighiero G., et al., 2014, arXiv.org, pp 19–30Rosales-Ortega F. F., S´anchez S. F., Iglesias-P´aramo J., D´ıazA. I., V´ılchez J. M., Bland-Hawthorn J., Husemann B., MastD., 2012, ApJ, 756, L31Saintonge A., et al., 2016, arXiv.org, pp 1749–1756S´anchez S. F., et al., 2012, Astronomy & Astrophysics, 538, A8S´anchez S. F., et al., 2013, A&A, 554, A58Santini P., et al., 2009, A&A, 504, 751Santini P., et al., 2017, ApJ, 847, 76Sargent M. T., Bethermin M., Daddi E., Elbaz D., 2012, TheAstrophysical Journal Letters, 747, L31Schreiber C., et al., 2015, Astronomy & Astrophysics, 575, A74Schruba A., Leroy A. K., Walter F., Sand strom K., RosolowskyE., 2010, ApJ, 722, 1699Shivaei I., Reddy N. A., Steidel C. C., Shapley A. E., 2015, ApJ,804, 149Simons R. C., et al., 2017, ApJ, 843, 46Skrutskie M. F., et al., 2006, AJ, 131, 1163Smith D. J. B., Hayward C. C., 2015, MNRAS, 453, 1597Smith D. J. B., Hayward C. C., 2018, MNRAS, 476, 1705Smith D. J. B., et al., 2012, MNRAS, 427, 703Sobral D., Best P. N., Smail I., Mobasher B., Stott J., Nisbet D.,2014, MNRAS, 437, 3516MNRAS000
Brammer G., et al., 2012, arXiv.org, p. 13Bruzual G., Charlot S., 2003, Monthly Notices of the Royal As-tronomical Society, 344, 1000Bundy K., et al., 2015, The Astrophysical Journal, 798, 7Calzetti D., et al., 2007, ApJ, 666, 870Camps P., Baes M., 2015, Astronomy and Computing, 9, 20Cano-D´ıaz M., et al., 2016, The Astrophysical Journal Letters,821, L26Cano-D´ıaz M., ´Avila-Reese V., S´anchez S. F., Hern´andez-ToledoH. M., Rodr´ıguez-Puebla A., Boquien M., Ibarra-Medel H.,2019, MNRAS, 488, 3929Casasola V., Hunt L., Combes F., Garc´ıa-Burillo S., 2015, A&A,577, A135Chabrier G., 2003, The Publications of the Astronomical Societyof the Pacific, 115, 763Chang Y.-Y., van der Wel A., da Cunha E., Rix H.-W., 2015,ApJS, 219, 8Charlot S., Fall S. M., 2000, ApJ, 539, 718Chevance M., et al., 2019, arXiv e-prints, p. arXiv:1911.03479Chrimes A. A., Stanway E. R., Levan A. J., Davies L. J. M.,Angus C. R., Greis S. M. L., 2018, MNRAS, 478, 2Clark C. J. R., et al., 2018, A&A, 609, A37Conselice C. J., Wilkinson A., Duncan K., Mortlock A., 2016, TheAstrophysical Journal, 830, 83Corwin Harold G. J., Buta R. J., de Vaucouleurs G., 1994, AJ,108, 2128Croom S. M., et al., 2012, Monthly Notices of the Royal Astro-nomical Society, 421, 872Daddi E., et al., 2007, The Astrophysical Journal, 670, 156Davies J. I., et al., 2017, PASP, 129, 044102Driver S. P., et al., 2018, MNRAS, 475, 2891Elbaz D., et al., 2007, Astronomy & Astrophysics, 468, 33Foreman-Mackey D., Hogg D. W., Lang D., Goodman J., 2013,PASP, 125, 306F¨orster Schreiber N. M., et al., 2009, ApJ, 706, 1364F¨orster Schreiber N. M., et al., 2018, ApJS, 238, 21Glazebrook K., 2013, Publ. Astron. Soc. Australia, 30, e056Gonz´alez Delgado R. M., et al., 2016, A&A, 590, A44Grogin N. A., et al., 2011, ApJS, 197, 35 Hall C., Courteau S., Jarrett T., Cluver M., Meurer G., CarignanC., Audcent-Ross F., 2018, ApJ, 865, 154Hayward C. C., Smith D. J. B., 2015, MNRAS, 446, 1512Hemmati S., et al., 2014, ApJ, 797, 108Hsieh B. C., et al., 2017, ApJ, 851, L24Jafariyazani M., Mobasher B., Hemmati S., Fetherolf T., Khos-tovan A. A., Chartab N., 2019, ApJ, 887, 204Kennicutt Robert C. J., 1998, ApJ, 498, 541Khoperskov S. A., Vasiliev E. O., 2017, MNRAS, 468, 920Koekemoer A. M., et al., 2011, The Astrophysical Journal Sup-plement, 197, 36Kravtsov A. V., 2003, ApJ, 590, L1Law D. R., Steidel C. C., Erb D. K., Larkin J. E., Pettini M.,Shapley A. E., Wright S. A., 2009, ApJ, 697, 2057Leroy A. K., Walter F., Brinks E., Bigiel F., de Blok W. J. G.,Madore B., Thornley M. D., 2008, AJ, 136, 2782Lin L., et al., 2017, ApJ, 851, 18Magdis G. E., et al., 2016, Monthly Notices of the Royal Astro-nomical Society, 456, 4533Makarov D., Prugniel P., Terekhova N., Courtois H., Vauglin I.,2014, A&A, 570, A13Maragkoudakis A., Zezas A., Ashby M. L. N., Willner S. P., 2017,Monthly Notices of the Royal Astronomical Society, 466, 1192Martis N. S., Marchesini D. M., Muzzin A., Stefanon M., Bram-mer G., da Cunha E., Sajina A., Labbe I., 2019, ApJ, 882,65Medling A. M., et al., 2018, Monthly Notices of the Royal Astro-nomical Society, 475, 5194Meurer G. R., et al., 2006, ApJS, 165, 307Morrissey P., et al., 2007, ApJS, 173, 682Noeske K. G., et al., 2007, The Astrophysical Journal, 660, L47Oliver S., et al., 2010, MNRAS, 405, 2279Pannella M., et al., 2009, arXiv.org, pp L116–L120Pearson W. J., et al., 2018, A&A, 615, A146Peng Y., Maiolino R., Cochrane R., 2015, Nature, 521, 192P´erez E., et al., 2013, ApJ, 764, L1Pilbratt G. L., et al., 2010, A&A, 518, L1Planck Collaboration et al., 2016, A&A, 594, A13Poggianti B. M., et al., 2017, ApJ, 844, 48Popesso P., et al., 2019a, MNRAS, 483, 3213Popesso P., et al., 2019b, MNRAS, 490, 5285Renzini A., Peng Y.-j., 2015, The Astrophysical Journal Letters,801, L29Rodighiero G., et al., 2011, The Astrophysical Journal Letters,739, L40Rodighiero G., et al., 2014, arXiv.org, pp 19–30Rosales-Ortega F. F., S´anchez S. F., Iglesias-P´aramo J., D´ıazA. I., V´ılchez J. M., Bland-Hawthorn J., Husemann B., MastD., 2012, ApJ, 756, L31Saintonge A., et al., 2016, arXiv.org, pp 1749–1756S´anchez S. F., et al., 2012, Astronomy & Astrophysics, 538, A8S´anchez S. F., et al., 2013, A&A, 554, A58Santini P., et al., 2009, A&A, 504, 751Santini P., et al., 2017, ApJ, 847, 76Sargent M. T., Bethermin M., Daddi E., Elbaz D., 2012, TheAstrophysical Journal Letters, 747, L31Schreiber C., et al., 2015, Astronomy & Astrophysics, 575, A74Schruba A., Leroy A. K., Walter F., Sand strom K., RosolowskyE., 2010, ApJ, 722, 1699Shivaei I., Reddy N. A., Steidel C. C., Shapley A. E., 2015, ApJ,804, 149Simons R. C., et al., 2017, ApJ, 843, 46Skrutskie M. F., et al., 2006, AJ, 131, 1163Smith D. J. B., Hayward C. C., 2015, MNRAS, 453, 1597Smith D. J. B., Hayward C. C., 2018, MNRAS, 476, 1705Smith D. J. B., et al., 2012, MNRAS, 427, 703Sobral D., Best P. N., Smail I., Mobasher B., Stott J., Nisbet D.,2014, MNRAS, 437, 3516MNRAS000 , 1–14 (2020) Enia A. et al.
Solomon P. M., Rivolo A. R., Barrett J., Yahil A., 1987, ApJ,319, 730Speagle J. S., Steinhardt C. L., Capak P. L., Silverman J. D.,2014, arXiv.org, p. 15Steinhardt C. L., et al., 2014, The Astrophysical Journal Letters,791, L25Tacchella S., Dekel A., Carollo C. M., Ceverino D., DeGraf C.,Lapiner S., Mandelker N., Primack J. R., 2015, arXiv.org, pp2790–2813¨Ubler H., et al., 2019, ApJ, 880, 48Viaene S., et al., 2014, A&A, 567, A71Vulcani B., et al., 2019, MNRAS, 488, 1597Werner M. W., et al., 2004, ApJS, 154, 1Whitaker K. E., et al., 2014, The Astrophysical Journal, 795, 104Williams T. G., Gear W. K., Smith M. W. L., 2018, MNRAS,479, 297Wisnioski E., et al., 2015, ApJ, 799, 209Wright E. L., et al., 2010, AJ, 140, 1868Wuyts S., et al., 2013, The Astrophysical Journal, 779, 135York D. G., et al., 2000, AJ, 120, 1579da Cunha E., Charlot S., Elbaz D., 2008, MNRAS, 388, 1595da Cunha E., Charmandaris V., D´ıaz-Santos T., Armus L., Mar-shall J. A., Elbaz D., 2010, A&A, 523, A78de Vaucouleurs G., de Vaucouleurs A., Corwin Herold G. J., ButaR. J., Paturel G., Fouque P., 1991, Third Reference Catalogueof Bright Galaxies
APPENDIX A: GALAXY-BY-GALAXYRESULTS
In this Appendix, we present the individual galaxy-by-galaxy results, reporting the SFR and stellar mass maps,along with a map measuring the distance from the fittedmain sequence (blue points towards starburst, red pointstowards quenched ones). In the lower left panel, we reportthe results on the M (cid:63) -SFR plane, with fitted (total) MS inblack and the one fitting the lone galaxy results. This paper has been typeset from a TEX/L A TEX file prepared bythe author. MNRAS , 1–14 (2020) . Sub-kpc MS in nearby spirals ◦
39’ 36’ 33’ 30’41 ◦ RA [deg] D E C [ deg ] − . − . − . − . − . − . log SFR[ M (cid:12) /yr ] ◦
39’ 36’ 33’ 30’41 ◦ RA [deg] D E C [ deg ] − . − . − . − .
25 0 .
00 0 .
25 0 .
50 0 .
75 1 . Distance from MS ◦
39’ 36’ 33’ 30’41 ◦ RA [deg] D E C [ deg ] . . . . . . . log M ∗ [ M (cid:12) ] . . . . . . . . . log Σ ∗ [M (cid:12) kpc − ] − . − . − . − . − . − . − . − . − . − . l og Σ S F R [ M (cid:12) y r − k p c − ] R = . . . . . . . . Distance from center [1 /R ] NGC3184, SABc
Figure A1.
Same as Fig. 4, for NGC3184MNRAS000
Same as Fig. 4, for NGC3184MNRAS000 , 1–14 (2020) Enia A. et al. ◦
16’ 14’ 12’ 10’ 08’44 ◦ RA [deg] D E C [ deg ] − . − . − . − . − . − . − . − . log SFR[ M (cid:12) /yr ] ◦
16’ 14’ 12’ 10’ 08’44 ◦ RA [deg] D E C [ deg ] − . − . − . − .
25 0 .
00 0 .
25 0 .
50 0 .
75 1 . Distance from MS ◦
16’ 14’ 12’ 10’ 08’44 ◦ RA [deg] D E C [ deg ] . . . . . . log M ∗ [ M (cid:12) ] . . . . . . . . . log Σ ∗ [M (cid:12) kpc − ] − . − . − . − . − . − . − . − . − . − . l og Σ S F R [ M (cid:12) y r − k p c − ] R = . . . . . . Distance from center [1 /R ] NGC3938, Sc
Figure A2.
Same as Fig. 4, for NGC3938. MNRAS , 1–14 (2020) . Sub-kpc MS in nearby spirals ◦
44’ 42’ 40’14 ◦ RA [deg] D E C [ deg ] − . − . − . − . − . − . − . log SFR[ M (cid:12) /yr ] ◦
44’ 42’ 40’14 ◦ RA [deg] D E C [ deg ] − . − . − . − .
25 0 .
00 0 .
25 0 .
50 0 .
75 1 . Distance from MS ◦
44’ 42’ 40’14 ◦ RA [deg] D E C [ deg ] . . . . . . . . log M ∗ [ M (cid:12) ] . . . . . . . . . log Σ ∗ [M (cid:12) kpc − ] − . − . − . − . − . − . − . − . − . − . l og Σ S F R [ M (cid:12) y r − k p c − ] R = . . . . . . . . Distance from center [1 /R ] NGC4254, Sc
Figure A3.
Same as Fig. 4, for NGC4254.MNRAS000
Same as Fig. 4, for NGC4254.MNRAS000 , 1–14 (2020) Enia A. et al. ◦
48’ 46’ 44’ 42’ 40’15 ◦ RA [deg] D E C [ deg ] − . − . − . − . − . − . − . − . log SFR[ M (cid:12) /yr ] ◦
48’ 46’ 44’ 42’ 40’15 ◦ RA [deg] D E C [ deg ] − . − . − . − .
25 0 .
00 0 .
25 0 .
50 0 .
75 1 . Distance from MS ◦
48’ 46’ 44’ 42’ 40’15 ◦ RA [deg] D E C [ deg ] . . . . . . . . log M ∗ [ M (cid:12) ] . . . . . . . . . log Σ ∗ [M (cid:12) kpc − ] − . − . − . − . − . − . − . − . − . − . l og Σ S F R [ M (cid:12) y r − k p c − ] R = . . . . . . Distance from center [1 /R ] NGC4321, SABb
Figure A4.
Same as Fig. 4, for NGC4321. MNRAS , 1–14 (2020) . Sub-kpc MS in nearby spirals ◦
38’ 36’ 34’ 32’8 ◦ RA [deg] D E C [ deg ] − . − . − . − . − . − . − . log SFR[ M (cid:12) /yr ] ◦
38’ 36’ 34’ 32’8 ◦ RA [deg] D E C [ deg ] − . − . − . − .
25 0 .
00 0 .
25 0 .
50 0 .
75 1 . Distance from MS ◦
38’ 36’ 34’ 32’8 ◦ RA [deg] D E C [ deg ] . . . . . . . . . log M ∗ [ M (cid:12) ] . . . . . . . . . log Σ ∗ [M (cid:12) kpc − ] − . − . − . − . − . − . − . − . − . − . l og Σ S F R [ M (cid:12) y r − k p c − ] R = . . . . . . . . . Distance from center [1 /R ] NGC4535, Sc
Figure A5.
Same as Fig. 4, for NGC4535.MNRAS000
Same as Fig. 4, for NGC4535.MNRAS000 , 1–14 (2020) Enia A. et al. ◦
35’ 30’ 25’ 20’47 ◦ RA [deg] D E C [ deg ] − . − . − . − . − . − . − . − . − . log SFR[ M (cid:12) /yr ] ◦
35’ 30’ 25’ 20’47 ◦ RA [deg] D E C [ deg ] − . − . − . − .
25 0 .
00 0 .
25 0 .
50 0 .
75 1 . Distance from MS ◦
35’ 30’ 25’ 20’47 ◦ RA [deg] D E C [ deg ] . . . . . . . . log M ∗ [ M (cid:12) ] . . . . . . . . . log Σ ∗ [M (cid:12) kpc − ] − . − . − . − . − . − . − . − . − . − . l og Σ S F R [ M (cid:12) y r − k p c − ] R = . . . . . . Distance from center [1 /R ] NGC5194, Sbc
Figure A6.
Same as Fig. 4, for NGC5194. MNRAS , 1–14 (2020) . Sub-kpc MS in nearby spirals ◦
00’ 210 ◦
50’ 40’54 ◦ RA [deg] D E C [ deg ] − . − . − . − . − . − . log SFR[ M (cid:12) /yr ] ◦
00’ 210 ◦
50’ 40’54 ◦ RA [deg] D E C [ deg ] − . − . − . − .
25 0 .
00 0 .
25 0 .
50 0 .
75 1 . Distance from MS ◦
00’ 210 ◦
50’ 40’54 ◦ RA [deg] D E C [ deg ] . . . . . . . . log M ∗ [ M (cid:12) ] . . . . . . . . . log Σ ∗ [M (cid:12) kpc − ] − . − . − . − . − . − . − . − . − . − . l og Σ S F R [ M (cid:12) y r − k p c − ] R = . . . . . . . Distance from center [1 /R ] NGC5457, SABc
Figure A7.
Same as Fig. 4, for NGC5457.MNRAS000