The Physical Characteristics of Interstellar Medium in NGC 3665 with Herschel Observations
TThe Physical Characteristics of Interstellar Medium in NGC 3665with
Herschel
Observations (cid:63)
Meng-Yuan Xiao , , , Yinghe Zhao , , , Qiu-Sheng Gu , , , Yong Shi , , School of Astronomy and Space Science, Nanjing University, Nanjing 210093,P. R. China; [email protected] Key Laboratory of Modern Astronomy and Astrophysics, Nanjing University, Nanjing210093, P. R. China Yunnan Observatories, CAS, Kunming 650011, P. R. China Key Laboratory for the Structure and Evolution of Celestial Objects, CAS, Kunming650011, P. R. China Collaborative Innovation Center of Modern Astronomy and Space Exploration, Nanjing210093, P. R. China Center for Astronomical Mega-Science, CAS, 20A Datun Road, Chaoyang District,Beijing 100012, P. R. China
ABSTRACT
We present the analysis of the physical properties of the interstellar medium(ISM) in the nearby early-type galaxy NGC 3665, based on the far-infrared (FIR)photometric and spectroscopic data as observed by the Herschel Space Observa-tory. The fit to the spectral energy distribution reveals a high dust content inthe galaxy, with the dust-to-stellar mass ratio of M dust / M ∗ ∼ × − thatis nearly three times larger than the mean value of local S0+S0a galaxies. Forthe ionized regions (H ii regions), the electron density (n e ) is around 49.5 ± − based on the [N ii ] 122 µ m/[N ii ] 205 µ m ratio. For the photodissociationregions, the heating efficiency is in the range of 1.26 × − and 1.37 × − based on the ([C ii ] + [O i ] 63 µ m)/ L TIR , which is slightly lower than other localgalaxies; the hydrogen nucleus density and the strength of FUV radiation field (cid:63)
Herschel is an ESA space observatory with science instruments provided by European-led PrincipalInvestigator consortia and with important participation from NASA. a r X i v : . [ a s t r o - ph . GA ] J a n ∼ cm − and G ∼ − . , respectively. The above results are consistentwith the presence of weak AGN and a low level of star-forming activity in NGC3665. Our results give strong support to the ‘morphological quenching’ scenario,where a compact, massive bulge can stabilize amount of cool gas against starformation. Subject headings: galaxies: individual (NGC 3665) – galaxies: elliptical andlenticular, cD – galaxies: ISM – infrared: ISM – ISM: lines and bands
1. Introduction
Interstellar medium (ISM) plays a crucial role in the galaxy formation and evolution. Itis the primary reservoir for star formation. The mutual interaction between ISM and starsdetermines the rates of both gaseous depletion and star formation in the galaxy. ISM containsseveral main components: ionized gas, neutral gas, cold molecular clouds, as well as dustgrains. When molecular clouds collapse under their own gravity to form stars, the gravityis required to overcome the random motion pressure, which requires the interstellar gas tobe cooled down sufficiently. Neutral and ionized gas can be heated up by the photoelectricprocess (PE; Tielens & Hollenbach 1985), and cooled via collisional excitation of C + , O,N + and other elements. The atomic fine-structure emission lines in the far-infrared (FIR),such as [N ii ] 122 and 205 µ m, [C ii ] 158 µ m, [O i ] 63 µ m, and [C i ] 370 µ m, are veryimportant coolants, which play a crucial role in the thermal balance in the H II regions andphotodissociation regions (PDRs), and can be served as critical diagnostic tools for studyingthe physical properties of ISM (e.g., Kaufman et al. 1999).Among these fine-structure lines, [C ii ] 158 µ m is the brightest and the dominant FIRcooling line, typically accounting for 0.1%-1% of the total FIR luminosity (Stacey et al.1991; Malhotra et al. 2001; D´ıaz-Santos et al. 2013; Sargsyan et al. 2014). The ionizationpotential of carbon is 11.26 eV, thus the [C ii ] line can be a tracer of both ionized andneutral gas. The [O i ] emission originates in the PDRs since the ionization potential ofoxygen is just above 13.6 eV, whereas the [N ii ] line exclusively arises from the ionized gasas nitrogen has an ionization potential of 14.5 eV. As shown in previous studies (Zhao etal. 2013, 2016a; Sargsyan et al. 2014), both [N ii ] 205 µ m and [C ii ] 158 µ m lines can beserved as useful indicators of star formation rate (SFR). Based on the critical densities ofcollisional excitations for [N ii ] 122 and 205 µ m, ∼
293 cm − and 44 cm − at the electrontemperature of 8000 K, respectively, the [N ii ] 122/[N ii ] 205 ratio is a sensitive probe of theelectron density (10 (cid:46) n e (cid:46)
300 cm − ) of ionized gas, which can be further used to estimatethe fraction of [C ii ] 158 µ m originating from ionized gas (Oberst et al. 2006, 2011). 3 –Based on Infrared Space Observatory ( ISO ) observations, Malhotra et al. (2000) inves-tigated the physical properties of gas and dust in four elliptical/S0 galaxies, and proposedthat softer radiation field might result in lower [C ii ]/ F FIR ratios than those of normal star-forming galaxies by a factor of 2 −
5. With data of
Herschel Space Observatory (hereafter
Herschel ; Pilbratt et al. 2010), Lapham et al. (2017) presented spectroscopic observationsof the FIR emission lines in 20 nearby elliptical/S0 galaxies and found that the average of[C ii ]/ F FIR ratio is slightly lower than that of spiral galaxies, and [C ii ] luminosity can beserved as a good SFR tracer for both early-type galaxies (ETGs) and spirals. Furthermore,Lapham et al. (2017) showed that the fraction of [C II] emission arising from ionized gas issimilar in ETGs (63.5 %) and normal spirals (53.0 %).In this work, we investigate the ISM properties of a nearby ETG, NGC 3665. In general,ETGs in the local Universe contain very little cool gas and/or dust. Many studies focused onthe mechanisms of star formation suppression for nearby ETGs, such as removing cool gas(Di Matteo et al. 2005; Hopkins et al. 2006), suppressing gas infall and cooling (Birnboim& Dekel 2003; Croton et al. 2006; Lanz et al. 2016), or stabilizing gas reservoirs (Martig etal. 2009; Tacchella et al. 2015), etc. However, several recent works have shown that ∼ M (cid:12) yr − (Temi et al. 2009a,b).NGC 3665 (11 h m s .7, +38 ◦ (cid:48) (cid:48)(cid:48) ) is located 33.1 Mpc away (Cappellari et al. 2011),with an inclination angle, i , of 69.9 ◦ (Onishi et al. 2017), and Hubble type of SA0 (RC3; deVaucouleurs et al. 1991). It is found to be gas-rich in the center (Young et al. 2011; Serraet al. 2012; Alatalo et al. 2013; Davis et al. 2013, 2014; Kamenetzky et al. 2016; Nylandet al. 2016; Onishi et al. 2017), with a total molecular gas mass of 10 . ± . M (cid:12) from theCombined Array for Research in Millimeter Astronomy (CARMA) observations (Alatalo etal. 2013). Alatalo et al. (2013) showed a prominent dusty disk in the g − r image, and aregularly rotating central molecular gas disk in the CO (1 −
0) emission. Onishi et al. (2017)obtained similar results of a prominent dust structure in the
Hubble Space Telescope (HST) H -band image, and a centrally-concentrated gaseous disk in the CO(2 −
1) observations,and derived the mass of the central super-massive black hole of 5.75 × M (cid:12) . In Daviset al. (2014), NGC 3665 shows the largest deviation from the so-called Kennicutt-Schmidtrelation (Kennicutt 1998a) among 32 CO-detected ATLAS ETGs. The abundant cool gasand low SFR surface density (Σ
SFR ) suggest that NGC 3665 is an atypical ETG. Throughstudying specific ISM properties, we can better understand its relation to the star formation,especially the reasons for such a low star formation efficiency.In this paper, we focus on the photometric observations of NGC 3665 at 100, 160, 250,350 and 500 µ m, obtained with the Photodetector Array Camera and Spectrometer (PACS; 4 –Poglitsch et al. 2010) and the Spectral and Photometric Imaging REceiver (SPIRE; Griffinet al. 2010) onboard Herschel , with five fine-structure lines of [N ii ] 122 and 205 µ m, [C ii ]158 µ m, [O i ] 63 µ m, and [C i ] 370 µ m. We also perform optical spectroscopic observations ofNGC 3665 with the CAHA 3.5m telescope. Using these spectra combined with the analysisof multi-wavelength spectral energy distribution (SED), we investigate SFR, the gas heatingand cooling efficiency in PDRs, and the hydrogen nucleus density and the strength of FUVradiation field derived from the PDR models.This paper is organized as follows. Data reductions for Herschel photometric and spec-troscopic observations and CAHA 3.5m telescope spectroscopic observations are describedin Section 2. We present the results and discuss their implications in Section 3, including theclassification of nuclear activity, SED fitting, star formation rate, the ionized gas contribu-tion to the [C ii ] 158 µ m emission, the photoelectric heating efficiency of the interstellar gas,and some derived values using the PDR model. Then, we compare our results with respectto the star formation and gas to the previous works, and discuss the possible mechanismsto suppress star formation in NGC 3665 in Section 4. The conclusions for this work aresummarized in Section 5.
2. Observations and Data Reduction2.1.
Herschel
PACS and SPIRE Photometry
NGC 3665 was observed with
Herschel
PACS and SPIRE in two open time projectsGT2 mbaes 2 (PI: M. Baes) and OT1 lyoung 1 (PI: L. Young), respectively. We performedaperture photometry with the public Level 2 map products, which were downloaded from theHerschel Science Archive . To match the 36 (cid:48)(cid:48) resolution of the 500 µ m image, we convolvedother band images with the kernels provided in Gordon et al. (2008). We adopted theaperture of 40 (cid:48)(cid:48) radius around the source, and determined the sky values within 60-90 (cid:48)(cid:48) annulus surrounding the target galaxy. The photometric errors on the source were given bythe standard deviation of 15 annulus between 60 (cid:48)(cid:48) and 90 (cid:48)(cid:48) around the source, as well as the5% uncertainty of the fiducial stellar models in the PACS photometry and the confusion noisein the SPIRE photometry (Herschel Observers’ Manual , v.5.0.3, 2014). We then applied thecorresponding color and aperture corrections to these fluxes. The final photometric fluxesare given in Table 1. http://herschel.esac.esa.int/Science_Archive.shtml http://herschel.esac.esa.int/Docs/Herschel/html/Observatory.html Herschel
PACS and SPIRE Spectroscopy
The
Herschel
FIR spectroscopic observations of NGC 3665 were performed by the pro-gram OT1 lyoung 1 (PI: L. Young; Lapham et al. 2017), with a total exposure time of 60.7hours of
Herschel . We focused on the [N ii ] 122 and 205 µ m, [C i ] 370 µ m, [C ii ] 158 µ mand [O i ] 63 µ m fine-structure lines observed with the PACS (better than 12 (cid:48)(cid:48) ) and SPIRE( ∼ (cid:48)(cid:48) and ∼ (cid:48)(cid:48) ) instruments. The PACS integral-field spectrometer covers the 51-220 µ m range with a spectral reso-lution of ∼ − , which is composed of 5 × (cid:48)(cid:48) .
4. It covers a total projected field of view (FoV) of 47 (cid:48)(cid:48) × (cid:48)(cid:48) . The angularresolutions are 9 (cid:48)(cid:48) . ∼ µ m, 10 (cid:48)(cid:48) . ∼ µ m, and 11 (cid:48)(cid:48) . ∼ µ m. For NGC 3665,the [O i ] 63 µ m, [N ii ] 122 µ m and [C ii ] 158 µ m data are comprised of observations acquiredin the mapping mode. The Level 2 data with standard rebinned cubes, observed with linerange spectroscopy mode and reduced with the Herschel Interactive Processing Environment(HIPE; Ott 2010) version 14.2, were downloaded directly from the Herschel Science Archive.The spectrometer effective spectral resolutions are about 86 km s − (the third grating order),290 km s − (the first grating order) and 238 km s − (the first grating order) for [O i ] 63 µ m,[N ii ] 122 µ m and [C ii ] 158 µ m, respectively. The basic observational informations for eachline are summarized in Table 2.Following Farrah et al. (2013), we co-added the spectra of the central 9 spaxels tomeasure the line fluxes, due to the fact that the line-emitting regions in NGC 3665 arenot confined in the central spaxel according to its 100 micron continuum emission. Thenwe fitted the observed spectrum with two Gaussian functions (for the line) plus a linearcomponent (for the continuum), as shown in Fig. 1. The uncertainty of the integrated fluxwas calculated according to the rms of the continuum. A point-source aperture correction(Balog et al. 2014) was applied to derive the final flux. The intrinsic line width ( σ true )was obtained using σ true = (cid:112) σ − σ , where σ obs and σ inst are the observed line widthand instrumental spectral resolution, respectively. The [O i ] 63 µ m is not detected, andthus we calculated its upper limit using 3 σ times the instrumental spectral resolution at theconsidered wavelength. The line fluxes are given in Table 2. In Fig. 2, we show the [C ii ]158 µ m and [N ii ] 122 µ m integrated intensity maps, overlaid with the CO( J = 1 − survey (Alatalo et al. 2013). The mapsare created by projecting the rasters onto a common, regular spatial grid with 1 . (cid:48)(cid:48) and1 . (cid:48)(cid:48) pixel sizes, respectively, clearly shows that the neutral and ionized gas has extended 6 –structures and follow the CO(1 −
0) gas disk. Both of these two cooling lines are strongestat the center, and weaker outwards, suggesting a mount of [C ii ] 158 µ m lines generate fromionized gas, which emit the [N ii ] 122 µ m. The offset on the nucleus in each panel of [C ii ]158 µ m and [N ii ] 122 µ m lines compared with CO(1 −
0) contour is smaller than 2 (cid:48)(cid:48) andwithin the beam size. Since the non-detection of [O i ] 63 µ m, and the [N ii ] 122 µ m emissionis marginally resolved, we tried but failed to perform radial decomposition. Thus, we onlyconcentrated on the analysis of the integrated properties of NGC 3665. The SPIRE Fourier-Transform Spectrometer (FTS) consists of two bolometer detectorarrays, the SPIRE Short Wavelength Spectrometer Array (SSW) and the SPIRE Long Wave-length Spectrometer Array (SLW), covering overlapping bands of 191 − µ m and 294 − µ m, respectively. The observations were conducted in the single pointing mode with a highspectral resolution of 0.04 cm − (or 1.2 GHz in frequency space). We used the Level 2 dataproducts, which were reduced using the standard pipeline provided by the HIPE version14.0, along with the SPIRE calibration version 14.2.To obtain the integrated fluxes of [C i ] and [N ii ] 205 µ m lines, we followed the methodin Lu et al. (2017). For the [C i ] 370 µ m line, we adopted a pure Sinc function as (1) thisline is only partially resolved given the large instrumental resolution ( ∼
540 km s − ) at 370 µ m, and (2) the S/N is not high enough to use a Sinc-convolved-Gauss function. For the[N ii ] 205 µ m line, we adopted a Sinc-convolved-Gauss function as the velocity resolutionat ∼ µ m is about 300 km s − . We also used FTLinefitter (FTFitter; in version 1.9) tofit the observed spectra, and obtained similar results. Table 2 gives the line fluxes of [N ii ]205 µ m and [C i ] 370 µ m.The SPIRE beam size at 205 µ m is ∼ (cid:48)(cid:48) , and thus might not cover the total emissionof the [N ii ] line. To check this, we used the correlation between the [N ii ] 205 emission tototal infrared luminosity ratio ( L [N ii ]205 µ m /L TIR ; see Section 3.2 for the calculation of L TIR used here) and FIR color (Zhao et al. 2016a): log( L [N ii ]205 µ m /L TIR ) = − . − x − x − x , where x = log( f /f ), and f and f represent the flux densities at 70and 160 µ m, respectively. We obtained the total [N ii ] 205 emission is ∼ × − Wm − , consistent with the measured value within the uncertainties. Furthermore, the [N ii ]flux is 7.29 ± × − W m − after calibrating with the Semi Extended CorrectionTool (SECT) in Lapham et al. (2017). Therefore, the distribution of [N ii ] 205 µ m emission (cid:48)(cid:48) beam. The optical spectroscopic observations were carried out with the 3.5-m telescope atCalar Alto Observatory (CAHA , Almer´ıa, Spain) on May 26, 2017. We used the PPAKintegral-field unit (IFU) (Verheijen et al. 2004) at Potsdam Multi-Aperture Spectrometer(PMAS; Roth et al. 2005) instrument, containing 382 fibers with each of 2 . (cid:48)(cid:48) diameter anda hexagonal FoV of 74 (cid:48)(cid:48) × (cid:48)(cid:48) (Kelz et al. 2006). The V500 grating, which has a spectralresolution (FWHM) of 6 ˚A and wavelength coverage of 3745-7500 ˚A, was adopted. A three-pointing dithering scheme was used with exposure time of 900 seconds each. The typicalairmass was ∼ (cid:48)(cid:48) × (cid:48)(cid:48) (9 spaxels).To obtain the emission line fluxes, we followed the method of Tremonti et al. (2004) andBrinchmann et al. (2004) to model the stellar continuum with templates, which are generatedusing the popular synthesis code of Bruzual & Charlot (2003, BC03) . The template spectraare composed of ten different ages (0.005, 0.025, 0.1, 0.2, 0.6, 0.9, 1.4, 2.5, 5, 10 Gyr) andfour metallicities (0.004, 0.008, 0.017, and 0.05). For each metallicity, we performed a non-negative least square fit to obtain the best-fitting model spectrum using the 10 single-agepopulations, with the internal dust attenuation model of Charlot & Fall (2000). During thefitting process, each template is convolved with a stellar velocity dispersion from 0 to 200 kms − by a step size of 5 km s − . After subtracting the best-fitting stellar continuum model,we obtained the pure nebular emission line spectrum, and fitted each line with one Gaussiancomponent, including H β , H α , [O iii ] λ ii ] λ
3. Results and Discussion3.1. Spectral Classification
To identify the power source of the emission lines, we adopted the well-known BPTdiagnostic diagram (Baldwin, Phillips & Terlevich 1981; Veilleux & Osterbrock 1987). Herewe only focus on the central 3 (cid:48)(cid:48) × (cid:48)(cid:48) region (the IFU data will be fully used in the followingpaper for a large sample of S0 galaxies) in NGC 3665. In Fig. 3, we plotted [N ii ] λ α versus [O iii ] λ β flux ratios with S / N > (cid:48)(cid:48) × (cid:48)(cid:48) ), whereas the red star shows the entire 3 (cid:48)(cid:48) × (cid:48)(cid:48) region. As shown in Fig. 3,all of the observed points lie in the composite region, suggesting that the central region ofNGC 3665 is a mixture of star formation and a weak AGN, and is consistent with Ho et al.(1997).For the central 3 (cid:48)(cid:48) × (cid:48)(cid:48) region, the equivalent width of H α ( W H α ) is ∼ ii ]/H α ) is about -0.21, which are also consistent with the identification of weak AGN(i.e., log([N ii ]/H α ) > -0.4 and W H α between 3 and 6 ˚A) in the W H α versus [N ii ]/H α (WHAN) diagram (Cid Fernandes et al. 2011). Our results are consistent with Nyland etal. (2016), who detected two extended radio jets on scale of kilo parsecs in NGC 3665.Using the observed ratio of F H α /F H β , we can estimate the nebular extinction, A V,nebular (Cardelli et al. 1989), assuming an unreddened I H α /I H β of 2.86 from Osterbrock (1989), e.g., A V,nebular = 7 . × log ( F H α /F H β I H α /I H β ) . (1)We obtained that the nebular extinction, A V,nebular , for NGC 3665 is ∼ ± (cid:48)(cid:48) × (cid:48)(cid:48) region, Σ SFR , without considering the effect of weak AGN. Using Kennicutt (1998b)relation with extinction-corrected H α luminosity:SFR ( M (cid:12) yr − ) = 7 . × − L ( H α ) (ergs s − ) , (2)we derive logΣ SFR ∼ − . M (cid:12) yr − kpc − , which is lower than the mean value, − M (cid:12) yr − kpc − , of 45 star-forming S0 galaxies with the same method (Xiao et al. 2016). 9 –Since the [N ii ] 122 µ m emission is the strongest in the center (see Fig. 2), suggesting asimilar distribution of H ii regions (Zhao et al. 2016a), we conclude that the star formationis concentrated in the galactic center and the rate is low. To better understand the infrared properties of NGC 3665, such as the total infraredluminosity ( L TIR ; 8-1000 µ m as defined in Sanders & Mirabel 1996), dust temperature( T dust ) and dust mass ( M dust ), we used the code of Multi-wavelength Analysis of GalaxyPhysical Properties (MAGPHYS; da Cunha et al. 2008) to fit the observed SED of NGC3665. Besides the Herschel
PACS and SPIRE photometric results, we also compiled UV toFIR photometries from Galaxy Evolution Explorer (
GALEX ; Loubser & S´anchez-Bl´azquez2011), SDSS (Adelman-McCarthy et al. 2008), Two Micron All Sky Survey (2MASS; Jarrettet al. 2000), and Infrared Astronomical Satellite (
IRAS ; Moshir et al. 1990) catalogs. Themeasured fluxes are listed in Table 3.MAGPHYS is a simple model to interpret in a consistent way the emission from galaxiesat UV, optical and IR wavelengths in terms of their star formation histories and dust content,using a Bayesian fitting method. The library of model galaxy spectra are composed of twotypes of binary files: the ‘optical models’ tracing the emission from stellar populations ingalaxies, calculated using BC03 with initial mass function (IMF) from Chabrier (2003) andthe dust attenuation model described in Charlot & Fall (2000); the ‘infrared models’ tracingthe emission from dust, following the approach described in da Cunha et al. (2008). Thismodel relies on the assumption that the total energy absorbed by dust from two maincomponents (Charlot & Fall 2000): the stellar birth clouds (star-forming regions) and theambient diffuse ISM, and re-radiated by dust at IR wavelengths via an energy balanceargument.We present the result of SED fitting for NGC 3665 in Fig. 4. In the top panel, the blackline shows the best model fitting to the observed data (red points), the blue and red linesrepresent the unattenuated stellar population spectrum and the dust emission, respectively.The residuals ( L obs − L mod ) /L obs ) are shown with black squares. As we can see that themodeled spectrum is in good agreement with the observed data points from GALEX
Herschel
SPIRE 500 µ m.Table 4 lists the derived parameters from the best-fit SED. The dust to stellar mass
10 –ratio M dust / M ∗ of NGC 3665 is about 1 . × − , which is about 3 times higher than themean value for 39 S0+S0a galaxies observed with Herschel (Smith et al. 2012). As shownin Alatalo et al. (2013), NGC 3665 has a total molecular gas mass of log M gas = 9.11 ± M (cid:12) , indicating that the gas-to-dust mass ratio (GDR) is ∼ ∼
120 from Li & Draine (2001), ∼
160 from Zubko et al. (2004), and ∼
180 from Draine et al. (2007). We also calculated the total infrared luminosity ( L IR ) ofNGC 3665, integrated within 8-1000 µ m from SED best-fitting model, to be 10 . ± . L (cid:12) .The error is estimated through performing a Monte Carlo simulation, sampling a series ofdata points according to a Gaussian distribution with the measured photometric values anderrors, and repeated the same fitting procedure for 1,000 times using the simulated datasets.We also ran another SED fitting model CIGALE (Noll et al. 2009) version 0.11.0 tomake a comparison. The derived values of stellar mass and dust luminosity are log M ∗ =10 . M (cid:12) and log L d = 9 . L (cid:12) , respectively, consistent with the results from MAGPHYS.Therefore, we adopted the fitted parameters from MAGPHYS in the following analysis. To estimate the star formation rate (SFR) in NGC 3665, we adopted several differentapproaches:(1) Using the infrared and
GALEX far-UV luminosity (Dale et al. 2007):SFR ( M (cid:12) yr − ) = 4 . × − L TIR (W) + 7 . × − νL ν (1500 ˚A) (W) , (3)where L TIR is calculated from integration within 8-1000 µ m from SED fitting (see the Section3.2). The SFR in NGC 3665 is derived to be 1.34 ± M (cid:12) yr − .(2) Based only on L TIR , with the algorithm of Kennicutt (1998b):SFR ( M (cid:12) yr − ) = 4 . × − L TIR (ergs s − ) . (4)The derived SFR is 1.29 ± M (cid:12) yr − . As we know, the weak AGN in NGC 3665 mightcontribute to the IR emission, which leads to an overestimated SFR. However, the results ofCIGALE suggest that the fractional contribution of AGN to the dust emission, f AGN , is lessthan 0.01, thus we ignored its contribution to the infrared luminosity.(3) As shown in Zhao et al. (2013, 2016a), the [N ii ] 205 µ m luminosity ( L [N ii ]205 µ m ) isless affected by emissions from older stars compared to the IR luminosity, it is also a good http://cigale.lam.fr
11 –indicator of SFR. With the 60-to-100 µ m flux density ratio, f /f ∼ . . ≤ f /f < . M (cid:12) yr − ) = − .
99 + log L [N ii ] ( L (cid:12) ) . (5)We calculated SFR to be 2.55 +1 . − . M (cid:12) yr − , where the error is estimated from the uncertainty(0.22 dex) associated with the SFR calibrator given in Zhao et al. (2016a).(4) The [C ii ] 158 µ m emission can also be served as a useful indicator of SFR (Sargsyanet al. 2014), as: log SFR ( M (cid:12) yr − ) = log L [C ii ] ( L (cid:12) ) − . , (6)with a scatter of 0.2 dex. The derived SFR is 1.66 +0 . − . M (cid:12) yr − .Therefore, SFRs from different calibrators are consistent with each other within uncer-tainties, and we take the averaged value of 1.7 M (cid:12) yr − as the final result. ii ] Emission from Ionized Gas The emission of [C ii ] originates from both the neutral and ionized gas, due to the lowionization potential (11.26 eV) of atomic carbon. To use the PDR models, in which onlythe emission from neutral gas has been taken into account, we first need to remove the[C ii ] emission from ionized gas. Following the method of Oberst et al. (2006, 2011), thecontribution of ionized gas can be estimated with the [C ii ]/[N ii ] 205 ratio, which is only afunction of electron density ( n e ) in the H ii regions after assuming a C/N abundance ratio. n e can be estimated with the [N ii ] 122/[N ii ] 205 ratio because of their different criticaldensities (Oberst et al. 2006; Zhao et al. 2016a). Compared the observed value (1 . ± . n e = 49.5 ± − , which is comparable tothose found in ETGs and star-forming galaxies. For instance, Lapham et al. (2017) found n e = 24 cm − for 11 nearby ETGs from Herschel observations, and D´ıaz-Santos et al. (2017)found n e from 20 to 100 cm − , with the mean value of 45 cm − , for 240 GOALS luminous IRgalaxies (LIRGs). In the Milky Way, the average value is measured to be 29 cm − (Goldsmithet al. 2015), and in nearby spiral galaxy NGC 891, n e is ranging from 1.9 to 80 cm − , witha mean value of 22 cm − (Hughes et al. 2015).Based on the Lick indices (Fe5015 and Mg b ) and single stellar population models,McDermid et al. (2015) derived the stellar metallicity at R e / ∼ ± ii ] λ α line ratio 12 –(Kewley & Dopita 2002) for the same central region, to be ∼ Z (cid:12) . Thus, by adoptinga solar abundances of C / H = 1 . × − and N/H = 7.9 × − from Savage & Sembach(1996), we further used the derived electron density to predict the [C ii ]/[N ii ] 205 ratioin ionized gas, and then compared it with the observed values. We find that the fractionof [C ii ] emission from ionized gas is about 43%. This value appears consistent with theprevious results from various sources that the majority of [C ii ] emission comes from PDRs(Abel et al. 2005; Oberst et al. 2006, 2011; Farrah et al. 2013; Parkin et al. 2014; Hughes etal. 2015; Lapham et al. 2017). After removing contribution from ionized gas for the [C ii ]emission, we estimated the [C ii ] flux originating from neutral gas is (27 . ± . × − Wm − . The [C ii ] 158 µ m and [O i ] 63 µ m are dominant coolants in neutral gas of PDRs, whichcan help us constrain the physical conditions of neutral ISM. The strengths of these two linesshow how many interstellar UV photons heat the gas by photoelectric effect, which can betraced by emission lines during gas cooling via collisional excitation at the FIR wavelengths.Another proportion of UV photons are absorbed by dust grains, and re-emit in the infrared,which can be traced by total infrared flux. Therefore, the ratio of ([C ii ]+[O i ]63)/ F TIR isa criterion for diagnosing photoelectric heating efficiency of the interstellar gas (Tielens &Hollenbach 1985).The total infrared flux, F TIR , is 2 . × − W m − , we calculated the ([C ii ]+[O i ]63)/ F TIR ratio in PDRs to be in the range of 1.26 × − and 1.37 × − , where the lower and up-per limits were obtained by assuming zero and 3 σ fluxes of the [O i ] emission, respectively.Whereas the typical values of ([C ii ]+[O i ]63)/ F TIR are in the range of 10 − to 10 − , both inETGs and late-type galaxies (Malhotra et al. 2001; Brauher et al. 2008; Lapham et al. 2017).For other galaxies with spatially resolved observations, the heating efficiency varies between ∼ × − and 10 − in the late spiral galaxy NGC 1097 and Seyfert 1 galaxy NGC 4559(Croxall et al. 2012). Hughes et al. (2015) also found in NGC 891, the ([C ii ]+[O i ]63)/ F TIR ranging from ∼ × − to 2 × − using Herschel
FIR spectroscopic observations. Parkinet al. (2014) showed that the heating efficiency in the disk of Centaurus A is ranging from 4 × − to 8 × − . In the arm and inter-arm regions of M51, the average value is up to ∼ − , and in the nucleus decreasing to 3 × − (Parkin et al. 2013). Therefore, NGC 3665is among those sources having the lowest gas heating efficiency.The low gas heating efficiency in NGC 3665 might be caused by its weak UV radiationfield. The gas is mainly photoelectrically heated by the UV photons, while dust can be 13 –heated by both optical and UV photons (Malhotra et al. 2000). As shown in Abel et al.(2009), the FIR color is a good indicator of the ionization parameter ( U ) of the ambient UVradiation field. For NGC 3665, f /f ∼ . U of ∼ − . Therefore,there is not enough energy for the electron to collisionally excited the C + and O to higherlevels. We compare IR emission line ratios to PDR model to obtain the physical properties ofthe PDR regions. Here we adopt the PDR model of Kaufman et al. (1999, 2006), which hasbeen updated based on the original model of Tielens & Hollenbach (1985). These modelsassume a homogeneous semi-infinite two-dimensional slab of a PDR and solve for the chem-istry, thermal balance, and radiation transfer simultaneously. For given gas-phase elementalabundances and grain properties, the model is parameterized by two free parameters: thehydrogen nucleus density, n , and the strength of FUV (6 eV < E < G , in units of 1.6 × − erg cm − s − from the local Galactic interstellar FUV field (Habing1968).We adopt the diagnostic observed line ratios of [C ii ]/[O i ]63 versus ([C ii ]+[O i ]63)/ F TIR as mentioned in Wolfire et al. (1990). Here we make several corrections to the observedquantities following the strategy of Zhao et al. (2016b). The cloud is optically thin to theinfrared continuum photon, which contributes to the actual observations from the front andback side of the cloud (especially when they are illuminated from all sides), while the modelsonly take into account one side emission exposed to the source of UV photons. Therefore,we reduce the observed F TIR by a factor of 2 as suggested by Kaufman et al. (1999).In the PDR models, the emission line is only considered to originate from neutral gas. Asmentioned above, the [C ii ] emission arises from both neutral and ionized gas, we first needto remove the contribution of [C ii ] from ionized gas, with the fraction of ∼ ii ] emission. The [C ii ] ismarginally optically thick with optical depth τ ∼ i ]63 line is optically thick. We observe the emission only from the front side of thecloud, while the other about half of the total [O i ]63 emission radiates away from the lineof sight. Accordingly, the actual observed [O i ]63 flux follows the geometrical assumption ofPDR models, without any correction applied. Finally, the equation of ([C ii ]+[O i ]63)/ F TIR
14 –after correcting is listed in here: ([C ii ]/1.4+[O i ]63)/( F TIR /2.0), where the [C ii ] emissionis only taken into account originating from neutral gas.Through a comparison of two observed line ratios to the two-dimensional PDR model,with χ minimization in the web-based Photo Dissociation Region Toolbox (PDRT; Pound& Wolfire 2008) , the derived values of hydrogen volume density, n , and the incident FUVradiation field, G , are listed in Table 5. We also list the best-fitting results derived fromuncorrected diagnostic observed line ratios in the PDR region compared to the model. Mean-while, both results before and after correction are shown in Fig. 5. The G in NGC 3665is significantly lower than what found in normal, star-forming, and starburst galaxies, andgalaxies with strong AGN (Negishi et al. 2001; Malhotra et al. 2001; Kramer et al. 2005;Oberst et al. 2011; Croxall et al. 2012; Parkin et al. 2013; Zhao et al. 2016b). The low G indicates a weak FUV radiation field, which is consistent with our previous analysis. F TIR is calculated among the whole galaxy, while [C ii ] emission tend to concentrate inthe center of NGC 3665 (see the Fig. 2) as well as the [O i ]63 emission, which might resultin the low G after comparing with the PDR model. Here we focus on the central spaxel,with a size of 9 (cid:48)(cid:48) . × (cid:48)(cid:48) .
4, to discuss the physical properties of ISM. The [C ii ] and [O i ]63line fluxes in the central region are measured using the same method as in Section 2.2.1. Wecalculate the central infrared luminosity of NGC 3665 using Herschel photometry at 100,160 and 250 µ m, following Galametz et al. (2013):L TIR ( L (cid:12) ) = (1 . ± . L µm + (0 . ± . L µm + (1 . ± . L µm , (7)where L µm , L µm and L µm are band luminosities in the unit of L (cid:12) . The centralinfrared luminosity is estimated to be (49.5 ± × L (cid:12) . After comparing the twoobserved line ratios to the PDR model, we derive two physical parameters in the centralPDRs of NGC 3665: the hydrogen nucleus density, n ∼ × cm − , and the strengthof FUV radiation field, G ∼ . . The G in the central region of NGC 3665 is about 3times larger than that derived from the whole galaxy, while this value is still low enough toindicate a weak FUV radiation field.
4. Abundant Molecular Gas and Suppressed Star Formation
Comparing with the so-called ‘star-forming main sequence’ (stellar mass-SFR relation)at z ∼ http://dustem.astro.umd.edu/pdrt/
15 –at the fixed stellar mass, showing that star formation is suppressed. In Davis et al. (2014),NGC 3665 has log Σ
SFR down to − . M (cid:12) yr − kpc − , and shows the largest deviationfrom the Kennicutt-Schmidt (KS) relation among 32 CO-detected ATLAS ETGs. Thegas (atomic+molecular) surface density (Σ gas ) of NGC 3665 is ∼ SFR . Furthermore, its molecular gas surface density (2.16 M (cid:12) pc − ; Davis et al. 2014) is significant larger than spiral galaxies at a given Σ SFR in theKS plane (Bigiel et al. 2008; Leroy et al. 2008, 2013). Shi et al. (2011, 2018) proposed anextended Schmidt law, invoking the stellar mass to be a secondary role (the first is gas mass)in regulating star formation, as Σ
SFR ∝ (Σ . Σ gas ) . . We also compared NGC 3665 with theextended Schmidt law, and found that it has the largest offset to this relation among 20ETGs, and has Σ . Σ gas ∼ SFR . Herewe have calculated the stellar mass surface density to be log Σ star = 3.56 M (cid:12) pc − with thesame method mentioned in the following as Fang et al. (2013). These results reveal thatNGC 3665 has large gas reservoirs while less star formation.Meanwhile, NGC 3665 is in a low density environment, with the local galaxy surfacedensity within the radius to the 10th nearest neighbor of log Σ = − . M pc − , andhas been classified into field galaxy in Cappellari et al. (2011). Young et al. (2011) furtherexplained that the poor environment might induce in large CO storage in galaxies by coolgas accretion. We derived the contribution from cold dust to the total dust luminosity ( ξ totC )to be ∼
70 per cent from MAGPHYS, and calculate f /f ∼ . Galfit (version3.0.5; Peng et al. 2002, 2010), and find that the SDSS (Gunn et al. 1998; Aihara et al. 2011) r -band image of NGC 3665 can be fitted very well with one S´ersic component. The fittingresults are listed in Table 6, and shown in Fig. 6. The S´ersic index is 3.81, showing thatNGC 3665 is a typical bulge-dominated galaxy. Fang et al. (2013) found the stellar masssurface density within 1 kpc, Σ , is a critical indicator of the star formation suppression.We calculate Σ following Fang’s method. Using the relation between stellar mass-to- i -bandluminosity ratio and rest-frame g − i color, log M/L i = 1 .
15 + 0 .
79 ( g − i ), we measure thevalue of log Σ to be ∼ M (cid:12) kpc − . This value is about 0.15 dex higher than the best-fitΣ vs. stellar mass relation for green valley and red sequence galaxies at the fixed stellar mass(Figure 4 in Fang et al. 2013), indicating a relatively compact spheroidal stellar componentin our source. With such a compact, massive bulge to stabilize cold gas reservoirs, star 16 –formation can be suppressed effectively in NGC 3665.On the other hand, we estimate the Toomre Q parameter to explore whether the molecu-lar gas disk is stable enough to against the gravitational fragmentation, following the criteriaof Toomre (1964): Q gas = σ gas κπG Σ gas > . (8) σ gas is the molecular gas velocity dispersion to be 12.53 km s − with the CARMA observations(Onishi et al. 2017), κ is the epicyclic frequency adopted of a approximate relation κ ∼√ V (R)/R, G is Newton’s gravitational constant, and Σ gas is the surface density of the gas diskto be ∼ M (cid:12) pc − . Thus we derive Q gas ∼ M (cid:12) , there is another possible mechanism to prevent star formationby shock heating the gas inflow from the halo (Kereˇs et al. 2005; Dekel & Birnboim 2006;Cappellari 2016). However, with the low percentage of hot gas, these two modes of starformation suppression are unlikely the dominant mechanisms in NGC 3665. The metallicityis ∼ Z (cid:12) , thus we can exclude the effect of the metallicity in lowing star formation (Shiet al. 2014).Consequently, the suppression of star formation in NGC 3665 is most possibly causedby its compact, massive bulge through stabilizing cold gas, which enables NGC 3665 to serveas a good observational sample for the stabilization of cold gas reservoirs, and is influencedby the ‘radio-mode’ feedback as well as the virial shocks. The low rate of star formation andweak AGN produce a weak UV radiation field (shown details in section 3.4.3), thus moredust is heated by old stars with inefficient photoelectric heating of the gas. The weak UVradiation field can not produce much [C ii ] and [O i ]63 emission and lead to somewhat low([C ii ]+[O i ]63)/ F TIR in this atypical early-type galaxy.
5. Conclusions
We present
Herschel
FIR photometric and spectroscopic observations of NGC 3665. Tobetter understand the nuclear activity in NGC 3665, we also conducted optical spectroscopicobservations. By combining the multi-wavelength data from literature and fitting the ob-served SED, we obtain dust luminosity, stellar and dust mass, dust temperature, infrared 17 –luminosity, and gas-to-dust mass ratio in NGC 3665. We discuss gas heating and coolingefficiency in the PDR regions and compare observed emission line ratios to the Kaufmanet al. (1999, 2006) PDR models to derive hydrogen nucleus density and strength of FUVradiation field. The main results are summarized as follows:1. From the PACS spectroscopic maps of [C ii ] 158 µ m and [N ii ] 122 µ m, we find thatboth neutral and ionized gas have extend structures and follow the CO(1 −
0) gas diskdistribution. The fluxes are strongest at the center, and gradually weaker outwards.2. NGC 3665 has dust-to-stellar mass ratio M dust /M ∗ ∼ . × − , which is nearly 3times larger than the mean value of local S0+S0a galaxies. The gas-to-dust mass ratiois 182, similar to that in the Milky Way, indicating a large gas reservoir.3. According to the BPT diagnostic diagram, NGC 3665 contains both star formation anda weak AGN in the central region. We calculated the SFR to be around 1.7 M (cid:12) yr − based on several different methods.4. The electron density of ionized gas in NGC 3665, based on the [N ii ] 122/[N ii ] 205ratio, is n e = 49 . ± . − . The contribution of ionized gas region to the total[C ii ] emission is about 43%, which is consistent with the previous results that themajority of [C ii ] emission comes from PDRs.5. The ([C ii ]+[O i ]63)/ F TIR line ratio is in the range of 1.26 × − and 1.37 × − ,indicating that NGC 3665 almost has the lowest gas heating efficiency in PDRs amongdifferent kinds of galaxies.6. A comparison between the observed emission line ratios and the theoretical PDR mod-els gives that the hydrogen nucleus density n ∼ . cm − , and the strength of FUVradiation field G ∼ − . , indicating a very weak UV radiation field in NGC 3665.7. After comparing our results with previous works, we find that NGC 3665 has largegas reservoirs while low-level star formation. The suppressed star formation is mostpossibly caused by its compact, massive bulge through stabilizing cool gas reservoirs. Acknowledgments
The authors are very grateful to the anonymous referee for critical comments and in-structive suggestions, which significantly strengthened the analyses in this work. We thankDr. Daizhong Liu and Dr. Zhiyu Zhang for helpful guidances about data reductions for 18 –
Herschel photometric and spectroscopic observations, thank Peng Wei for running SED fit-ting model CIGALE to help us compare the results with those in model MAGPHYS, andthank David Elbaz, Yifei Jin and Longji Bing for valuable discussions and advices whichimproved this paper. We are grateful to Sebasti´an S´anchez, resident astronomer at CAHA,for the optical spectroscopic observations and Rub´en Garc´ıa Benito for his help of data re-duction. This work is supported by the National Key Research and Development Program ofChina (No. 2017YFA0402703 and 2017YFA0402704), and by the National Natural ScienceFoundation of China (Nos. 11673057 and 11733002).PACS has been developed by a consortium of institutes led by MPE (Germany) andincluding UVIE (Austria); KU Leuven, CSL, IMEC (Belgium); CEA, LAM (France); MPIA(Germany); INAF-IFSI/OAA/OAP/OAT, LENS, SISSA (Italy); IAC (Spain). This de-velopment has been supported by the funding agencies BMVIT (Austria), ESA-PRODEX(Belgium), CEA/CNES (France), DLR (Germany), ASI/INAF (Italy), and CICYT/MCYT(Spain). SPIRE has been developed by a consortium of institutes led by Cardiff Univer-sity (UK) and including Univ. Lethbridge (Canada); NAOC (China); CEA, LAM (France);IFSI, Univ. Padua (Italy); IAC (Spain); Stockholm Observatory (Sweden); Imperial Col-lege London, RAL, UCL-MSSL, UKATC, Univ. Sussex (UK); and Caltech, JPL, NHSC,Univ. Colorado (USA). This development has been supported by national funding agencies:CSA (Canada); NAOC (China); CEA, CNES, CNRS (France); ASI (Italy); MCINN (Spain);SNSB (Sweden); STFC and UKSA (UK); and NASA (USA). HIPE is a joint development bythe Herschel Science Ground Segment Consortium, consisting of ESA, the NASA HerschelScience Center, and the HIFI, PACS and SPIRE consortia. Based on observations collectedat the Centro Astron´omico Hispano Alem´an (CAHA) at Calar Alto, operated jointly by theMax-Planck Institut f¨ur Astronomie and the Instituto de Astrof´ısica de Andaluc´ıa (CSIC).
REFERENCES
Abazajian, K. N., Adelman-McCarthy, J. K., Ag¨ueros, M. A., et al. 2009, ApJS, 182, 543Abel, N. P., Ferland, G. J., Shaw, G., & van Hoof, P. A. M. 2005, ApJS, 161, 65Abel, N. P., Dudley, C., Fischer, J., Satyapal, S., & van Hoof, P. A. M. 2009, ApJ, 701, 1147Adelman-McCarthy, J. K., Ag¨ueros, M. A., Allam, S. S., et al. 2008, ApJS, 175, 297-313Aihara, H., Allende Prieto, C., An, D., et al. 2011, ApJS, 193, 29Alatalo, K., Davis, T. A., Bureau, M., et al. 2013, MNRAS, 432, 1796 19 –Baldwin, J. A., Phillips, M. M., & Terlevich, R. 1981, PASP, 93, 5Balog, Z., M¨uller, T., Nielbock, M., et al. 2014, Experimental Astronomy, 37, 129Bigiel, F., Leroy, A., Walter, F., et al. 2008, AJ, 136, 2846Birnboim, Y., & Dekel, A. 2003, MNRAS, 345, 349Bluck, A. F. L., Mendel, J. T., Ellison, S. L., et al. 2014, MNRAS, 441, 599Brauher, J. R., Dale, D. A., & Helou, G. 2008, ApJS, 178, 280-301Brinchmann, J., Charlot, S., White, S. D. M., et al. 2004, MNRAS, 351, 1151Bruzual, G., & Charlot, S. 2003, MNRAS, 344, 1000Cappellari, M., Emsellem, E., Krajnovi´c, D., et al. 2011, MNRAS, 413, 813Cappellari, M. 2016, ARA&A, 54, 597Cardelli, J. A., Clayton, G. C., & Mathis, J. S. 1989, ApJ, 345, 245Chabrier, G. 2003, PASP, 115, 763Charlot, S., & Fall, S. M. 2000, ApJ, 539, 718Cid Fernandes, R., Stasi´nska, G., Mateus, A., & Vale Asari, N. 2011, MNRAS, 413, 1687Combes, F. 2017, Frontiers in Astronomy and Space Sciences, 4, 10Croton, D. J., Springel, V., White, S. D. M., et al. 2006, MNRAS, 365, 11Croxall, K. V., Smith, J. D., Wolfire, M. G., et al. 2012, ApJ, 747, 81da Cunha, E., Charlot, S., & Elbaz, D. 2008, MNRAS, 388, 1595Dale, D. A., Gil de Paz, A., Gordon, K. D., et al. 2007, ApJ, 655, 863Davis, T. A., Alatalo, K., Bureau, M., et al. 2013, MNRAS, 429, 534Davis, T. A., Young, L. M., Crocker, A. F., et al. 2014, MNRAS, 444, 3427de Vaucouleurs, G., de Vaucouleurs, A., Corwin, H. G., Jr., et al. 1991, Third ReferenceCatalogue of Bright Galaxies. Vol. I, Vol. II, Vol III (New York: Springer), 2091Dekel, A., & Birnboim, Y. 2006, MNRAS, 368, 2 20 –Di Matteo, T., Springel, V., & Hernquist, L. 2005, Nature, 433, 604D´ıaz-Santos, T., Armus, L., Charmandaris, V., et al. 2013, ApJ, 774, 68D´ıaz-Santos, T., Armus, L., Charmandaris, V., et al. 2017, arXiv:1705.04326Draine, B. T., Dale, D. A., Bendo, G., et al. 2007, ApJ, 663, 866Elbaz, D., Daddi, E., Le Borgne, D., et al. 2007, A&A, 468, 33Fang, J. J., Faber, S. M., Koo, D. C., & Dekel, A. 2013, ApJ, 776, 63Farrah, D., Lebouteiller, V., Spoon, H. W. W., et al. 2013, ApJ, 776, 38Galametz, M., Kennicutt, R. C., Calzetti, D., et al. 2013, MNRAS, 431, 1956Goldsmith, P. F., Yıldız, U. A., Langer, W. D., & Pineda, J. L. 2015, ApJ, 814, 133Gordon, K. D., Engelbracht, C. W., Rieke, G. H., et al. 2008, ApJ, 682, 336-354Griffin, M. J., Abergel, A., Abreu, A., et al. 2010, A&A, 518, L3Gunn, J. E., Carr, M., Rockosi, C., et al. 1998, AJ, 116, 3040Habing, H. J. 1968, Bull. Astron. Inst. Netherlands, 19, 421Ho, L. C., Filippenko, A. V., & Sargent, W. L. W. 1997, ApJS, 112, 315Hopkins, P. F., Hernquist, L., Cox, T. J., et al. 2006, ApJS, 163, 1Hughes, T. M., Foyle, K., Schirm, M. R. P., et al. 2015, A&A, 575, A17Jarrett, T. H., Chester, T., Cutri, R., et al. 2000, AJ, 119, 2498Kamenetzky, J., Rangwala, N., Glenn, J., Maloney, P. R., & Conley, A. 2016, ApJ, 829, 93Kaufman, M. J., Wolfire, M. G., Hollenbach, D. J., & Luhman, M. L. 1999, ApJ, 527, 795Kauffmann, G., Heckman, T. M., Tremonti, C., et al. 2003, MNRAS, 346, 1055Kaufman, M. J., Wolfire, M. G., & Hollenbach, D. J. 2006, ApJ, 644, 283Kelz, A., Verheijen, M. A. W., Roth, M. M., et al. 2006, PASP, 118, 129Kennicutt, R. C., Jr. 1998a, ApJ, 498, 541Kennicutt, R. C., Jr. 1998b, ARA&A, 36, 189 21 –Kereˇs, D., Katz, N., Weinberg, D. H., & Dav´e, R. 2005, MNRAS, 363, 2Kewley, L. J., Dopita, M. A., Sutherland, R. S., Heisler, C. A., & Trevena, J. 2001, ApJ,556, 121Kewley, L. J., & Dopita, M. A. 2002, ApJS, 142, 35Krajnovi´c, D., Emsellem, E., Cappellari, M., et al. 2011, MNRAS, 414, 2923Kramer, C., Mookerjea, B., Bayet, E., et al. 2005, A&A, 441, 961Lanz, L., Ogle, P. M., Alatalo, K., & Appleton, P. N. 2016, ApJ, 826, 29Lapham, R. C., Young, L. M., & Crocker, A. 2017, ApJ, 840, 51Leroy, A. K., Walter, F., Brinks, E., et al. 2008, AJ, 136, 2782Leroy, A. K., Walter, F., Sandstrom, K., et al. 2013, AJ, 146, 19Li, A., & Draine, B. T. 2001, ApJ, 554, 778Loubser, S. I., & S´anchez-Bl´azquez, P. 2011, MNRAS, 410, 2679Lu, N., Zhao, Y., D´ıaz-Santos, T., et al. 2017, ApJS, 230, 1Malhotra, S., Hollenbach, D., Helou, G., et al. 2000, ApJ, 543, 634Malhotra, S., Kaufman, M. J., Hollenbach, D., et al. 2001, ApJ, 561, 766Martig, M., Bournaud, F., Teyssier, R., & Dekel, A. 2009, ApJ, 707, 250McDermid, R. M., Alatalo, K., Blitz, L., et al. 2015, MNRAS, 448, 3484Moshir, M., et al. 1990, IRAS Faint Source Catalogue, version 2.0 (1990)Negishi, T., Onaka, T., Chan, K.-W., & Roellig, T. L. 2001, A&A, 375, 566Nesvadba, N. P. H., Boulanger, F., Salom´e, P., et al. 2010, A&A, 521, A65Noll, S., Burgarella, D., Giovannoli, E., et al. 2009, A&A, 507, 1793Nyland, K., Young, L. M., Wrobel, J. M., et al. 2016, MNRAS, 458, 2221Oberst, T. E., Parshley, S. C., Stacey, G. J., et al. 2006, ApJ, 652, L125Oberst, T. E., Parshley, S. C., Nikola, T., et al. 2011, ApJ, 739, 100 22 –Ogle, P., Antonucci, R., Appleton, P. N., & Whysong, D. 2007, ApJ, 668, 699Onishi, K., Iguchi, S., Davis, T. A., et al. 2017, arXiv:1703.05247Osterbrock, D. E. 1989, Astronomy, 17, 102Ott, S. 2010, Astronomical Data Analysis Software and Systems XIX, 434, 139Parkin, T. J., Wilson, C. D., Schirm, M. R. P., et al. 2013, ApJ, 776, 65Parkin, T. J., Wilson, C. D., Schirm, M. R. P., et al. 2014, ApJ, 787, 16Peng, C. Y., Ho, L. C., Impey, C. D., & Rix, H.-W. 2002, AJ, 124, 266Peng, C. Y., Ho, L. C., Impey, C. D., & Rix, H.-W. 2010, AJ, 139, 2097Pilbratt, G.L., Riedinger, J. R., Passvogel, T. et al. 2010, A&A, 518, L1Poglitsch, A., Waelkens, C., Geis, N., et al. 2010, A&A, 518, L2Pound, M. W., & Wolfire, M. G. 2008, Astronomical Data Analysis Software and SystemsXVII, 394, 654Roth, M. M., Kelz, A., Fechner, T., et al. 2005, PASP, 117, 620Sanders, D. B., & Mirabel, I. F. 1996, ARA&A, 34, 749Sargsyan, L., Samsonyan, A., Lebouteiller, V., et al. 2014, ApJ, 790, 15Savage, B. D., & Sembach, K. R. 1996, ARA&A, 34, 279Serra, P., Oosterloo, T., Morganti, R., et al. 2012, MNRAS, 422, 1835Shi, Y., Helou, G., Yan, L., et al. 2011, ApJ, 733, 87Shi, Y., Armus, L., Helou, G., et al. 2014, Nature, 514, 335Shi, Y., Yan, L., Armus, L., et al. 2018, arXiv:1801.00888Smethurst, R. J., Lintott, C. J., Simmons, B. D., et al. 2016, MNRAS, 463, 2986Smith, M. W. L., Gomez, H. L., Eales, S. A., et al. 2012, ApJ, 748, 123Stacey, G. J., Geis, N., Genzel, R., et al. 1991, ApJ, 373, 423Tacchella, S., Carollo, C. M., Renzini, A., et al. 2015, Science, 348, 314 23 –Temi, P., Brighenti, F., & Mathews, W. G. 2009, ApJ, 695, 1Temi, P., Brighenti, F., & Mathews, W. G. 2009, ApJ, 707, 890Temi, P., Brighenti, F., & Mathews, W. G. 2007a, ApJ, 660, 1215Temi, P., Brighenti, F., & Mathews, W. G. 2007b, ApJ, 666, 222Tielens, A. G. G. M., & Hollenbach, D. 1985, ApJ, 291, 722Toomre, A. 1964, ApJ, 139, 1217Tremonti, C. A., Heckman, T. M., Kauffmann, G., et al. 2004, ApJ, 613, 898Veilleux, S., & Osterbrock, D. E. 1987, ApJS, 63, 295Verheijen, M. A. W., Bershady, M. A., Andersen, D. R., et al. 2004, AstronomischeNachrichten, 325, 151Wolfire, M. G., Tielens, A. G. G. M., & Hollenbach, D. 1990, ApJ, 358, 116Xiao, M.-Y., Gu, Q.-S., Chen, Y.-M., & Zhou, L. 2016, ApJ, 831, 63Young, L. M., Bureau, M., Davis, T. A., et al. 2011, MNRAS, 414, 940Young, L. M., Scott, N., Serra, P., et al. 2014, MNRAS, 444, 3408Zhao, Y., Lu, N., Xu, C. K., et al. 2013, ApJ, 765, L13Zhao, Y., Lu, N., Xu, C. K., et al. 2016a, ApJ, 819, 69Zhao, Y., Yan, L., & Tsai, C.-W. 2016b, ApJ, 824, 146Zubko, V., Dwek, E., & Arendt, R. G. 2004, ApJS, 152, 211
This preprint was prepared with the AAS L A TEX macros v5.2.
24 –Table 1. A summary of the
Herschel
PACS and SPIRE Photometric Observations
Band Beam FWHM Size Pixel Size Color Correction Aperture Correction Flux( µm ) (arcsec) (arcsec) (Jy)100 7.7 1.6 1.000 1.271 6 . ± . . ± . . ± . . ± . . ± . Table 2. Fine-structure Lines Observed with the PACS and SPIRE Spectrometer
Line λ ObsID Obs Date Spec. Resolution Angular Resolution Flux FWHM( µm ) (km s − ) (arcsec) (10 − W m − ) (km s − )[O i ] P – P ∼ ∼ < · · · [N ii ] P – P ∼ ∼
10 14 . ± .
98 279, 221[C ii ] P / – P / ∼ ∼ . ± .
84 248, 251[N ii ] P – P ∼ ∼
17 7 . ± .
10 568[C i ] P – P ∼ ∼
36 0 . ± . · · · Note. — Column (1): atomic fine-structure line; columns (2): wavelength; columns (3): observation ID; column (4): observation date;column (5) and (6): spectral resolution and angular resolution from the PACS Observer’s Manual and SPIRE Handbook; column (7):measured fluxes of each atomic fine-structure line for NGC 3665; column (8): intrinsic line width.
Table 3. Photometry
Properties FUV NUV u g r i z (1) (2) (3) (4) (5) (6) (7) (8)Flux (mJy) 0 . ± .
02 1 . ± .
02 11 . ± .
03 58 . ± .
11 123 . ± .
23 187 . ± .
35 263 . ± . J H K s I µm I µm I µm I µm (9) (10) (11) (12) (13) (14) (15)567 . ± .
72 692 . ± .
04 564 . ± .
32 112 . ± .
10 162 . ± .
20 1920 . ± .
00 6340 . ± . GALEX far-UV (FUV) and near-UV (NUV) band, respectively;columns (4)-(8): flux of the SDSS u , g , r , i , and z band, respectively; columns (9)-(11): flux of the 2MASS J , H , and K s band, respectively;columns (12)-(15): fluxes of the four IRAS bands.
25 –Table 4. Derived properties from SED Fitting log M ∗ log L d log M d T BCW T ISMC log M gas M gas /M d log L TIR ( M (cid:12) ) ( L (cid:12) ) ( M (cid:12) ) (K) (K) ( M (cid:12) ) ( L (cid:12) )10.79 +0 . − . +0 . − . +0 . − . +3 . − . +0 . − . ± a
182 9.88 ± b Note. — The column from left to right are: stellar mass, dust luminosity, dust mass, temperatureof warm dust component in stellar birth clouds, temperature of cold dust component in diffuse ISM,total molecular gas mass, ratio of gas mass to dust mass, and total infrared luminosity. Uncertaintiesfrom MAGPHYS are the 16th-84th percentile range of the likelihood distribution from SED fitting. a The total molecular gas mass is from the CARMA observations (Alatalo et al. 2013). b The total infrared luminosity is integrated from SED best-fitting model between 8 and 1000 µ m,with the error obtained from a Monte Carlo simulation. Table 5. Results from the PDR model
Case log n log G (cm − ) (1.6 × − erg cm − s − )Uncorrected a b a The uncorrected values are derived from the bestfit, including the observed [C ii ] only from the neutralgas region, all observed [O i ]63 emission and F TIR . b The corrected values contain the [C ii ] divided bya factor of 1.4 in the neutral gas region and the F TIR reduced by 2.0.
Table 6. Results from the r -band bulge-disk decomposition Component m r R e S´ersic index b/a
PA chi / nu(mag) (arcsec) (deg)one S´ersic 11.18 50.15 3.81 0.78 28.12 0.95
26 – λ rest [ µ m]051015 F ν [ J y ] λ rest [ µ m]−20246810 62.8 62.9 63.0 63.1 63.2 63.3 63.4 λ rest [ µ m]−202468 [CII] 158 [NII] 122 [OI] 63 Fig. 1.— The [C ii ] 158 µ m, [N ii ] 122 µ m, and [O i ] 63 µ m spectra combined within central3 × i ] 63 µ m. 27 –Fig. 2.— The [C ii ] 158 µm and [N ii ] 122 µ m integrated intensity maps (color scale),overlaid with contours of CO(1 −
0) integrated intensity (moment0) from CARMA. We haveused the 3 σ cut to highlight the robust detections. Contour levels are 2.5, 16, 50, 84 per centof the peak, while the peak flux is 24.04 Jy beam − km s − . The color table on the right ofeach panel provides the integrated flux scale of [C ii ] 158 µm and [N ii ] 122 µ m, respectively,with the unit in 10 − W m − . The synthesized beam of Herschel and CARMA are shownin the bottom-left corner, with black open circle and grey filled circle, respectively. North isup, and east is to the left. 28 – −1.5 −1.0 −0.5 0.0 0.5log [NII] λ α −1.0−0.50.00.51.01.5 l og [ O III] λ / H β SF SeyfertLINER
Fig. 3.— The BPT diagram in the central 3 (cid:48)(cid:48) × (cid:48)(cid:48) spaxels. Green points are each individualspaxel, which lie between the red solid line (Kewley et al. 2001) and the red dash line andblue line (Kauffmann et al. 2003). Red star shows 9 spaxels as a whole. Both the greenpoints and red star, with the uncertainties, lie in the composite region in this diagnosticdiagram. 29 –Fig. 4.— Best MAGPHYS model fits (black line) to the observed SED (red points) ofthe NGC 3665. The data are composed of two GALEX , five SDSS, three 2MASS, three