SMA Discoveries in Nearby

12.2 NGC 3256 11.6 11.1 10.4 10 L⦿ 10 L⦿ 10 L⦿ NGC 253 10 L⦿

30", 5 kpc 2', 2 kpc Spitzer/GALEX HST/GOALS

Kazushi Sakamoto (ASIAA) in collaboration with S. Aalto, J. Black, F. Combes, J. Conway, F. Costagliola, A. Evans, P. Ho, D. Iono, E. Keto, R-Q. Mao, S. Martin, S. Matsushita, Y. Ohyama, A. Peck, G. Petitpas, M. Spaans, J. Wang, Z. Wang, M. Wiedner, D. Wilner, Q. Zhan, J-H. Zhao

SMA 10 @ Cambridge, MA, 2014 May 2004 Matsushita Submillimeter Array 12CO (J=3-2) Interferometric Observations of the Central Region of M51 2004 Sakamoto Molecular Gas around the Double Nucleus in M83 2004 Iono High-Density Molecular Gas in the Infrared-bright System VV 114 2004 Wang Warm Molecular Gas in Galaxy-Galaxy Merger NGC 6090 SMA on Nearby Galaxies 2005 Humphreys First Detection of Millimeter/Submillimeter Extragalactic H2O Maser Emission 2006 Sakamoto Molecular Superbubbles in the NGC 253 2006 Sakamoto Imaging Molecular Gas in the Luminous Merger NGC 3256: Detection of High-Velocity Gas and Twin Gas Peaks in the Double Nucleus 2007 Sakamoto Detection of CO Hot Spots Associated with Young Clusters in the Southern Starburst Galaxy NGC 1365 2007 Sawada-Satoh Structure and Kinematics of CO J = 2-1 Emission in the Central Region of NGC 4258 2007 Iono High-Resolution Imaging of Warm and Dense Molecular Gas in the Nuclear Region of the Luminous Infrared Galaxy NGC 6240 2007 Minh Unveiling the Ongoing Formation in the Starburst Galaxy NGC 253 54+ papers 2007 Chou The Circumnuclear Molecular Gas in the NGC 4945 2008 Ao Molecular Gas and in ARP 302 2008 Lim Radially Inflowing Molecular Gas in NGC 1275 Deposited by an X-Ray in the Cluster 920+ cites 2008 Hsieh Interferometric 12CO J = 2-1 Image of the Nuclear Region of Seyfert 1 Galaxy NGC 1097 2008 Papadopoulos A New Twist to an Old Story: HE 0450-2958 and the ULIRG --> Optically Bright QSO Transition Hypothesis 2008 Sakamoto Submillimeter Array Imaging of the CO(3-2) Line and 860 μm Continuum of Arp 220: Tracing the Spatial Distribution of Luminosity ~8 % of SMA 2008 Tan A Search for Molecular Gas in the Nucleus of M87 and Implications for the Fueling of Supermassive Black Holes as of June 4, 2014 2008 Wilson Luminous Infrared Galaxies with the Submillimeter Array. I. Survey Overview and the Central Gas to Dust Ratio 2009 Aalto High-resolution HNC 3-2 SMA observations of Arp 220 2009 Santangelo Resolving the molecular environment of super star clusters in Henize 2-10 excl. Time-domain, VLBI 2009 Matsushita SMA 12CO(J = 6 - 5) and 435 μm Interferometric Imaging of the Nuclear Region of Arp 220 2009 Iono Luminous Infrared Galaxies with the Submillimeter Array. II. Comparing the CO (3-2) Sizes and Luminosities of Local and High- Luminous Infrared Galaxies 2009 Muller HCO+ and HCN J = 3-2 Absorption Toward the Center of 2009 Ho Multiple Radial Cool Molecular Filaments in NGC 1275 2009 Sakamoto P Cygni Profiles of Molecular Lines Toward Arp 220 Nuclei 2010 Martin Imaging Carbon Monoxide Emission in the Starburst Galaxy NGC 6000 2010 Espada Disentangling the Circumnuclear Environs of Centaurus A. II. On the Nature of the Broad Absorption Line 2010 Sakamoto Vibrationally Excited HCN in the Luminous Infrared Galaxy NGC 4418 2011 Krips Submillimeter Array/Plateau de Bure Interferometer Multiple Line Observations of the Nearby Seyfert 2 Galaxy NGC 1068: Shock-related Gas Kinematics and Heating in the Central 100 pc? 2011 Boone High-resolution mapping of the physical conditions in two nearby active galaxies based on 12CO(1-0), (2-1), and (3-2) lines 2011 Martin The Submillimeter Array 1.3 mm line survey of Arp 220 2011 Lin The Central Region of the Nearby Seyfert 2 Galaxy NGC 4945: A Pair of Spirals 2011 Sakamoto Star-forming Cloud Complexes in the Central Molecular Zone of NGC 253 2011 Alatalo Discovery of an Driven Molecular Outflow in the Local Early-type Galaxy NGC 1266 2011 Hsieh Physical Properties of the Circumnuclear Starburst Ring in the Barred Galaxy NGC 1097 2011 Hirashita Central free-free dominated 880-μm emission in II Zw 40 2011 Li The alignment of molecular cloud magnetic fields with the spiral arms in M33 2012 Aalto Winds of change - a molecular outflow in NGC 1377?. The anatomy of an extreme FIR-excess galaxy 2012 Ueda Unveiling the Physical Properties and Kinematics of Molecular Gas in the (NGC 4038/9) through High-resolution CO (J = 3-2) Observations 2012 Tsai Interferometric CO(3-2) Observations toward the Central Region of NGC 1068 10 2012 Hsieh Probing Circumnuclear Environments with the HCN(J = 3-2) and HCO+(J = 3-2) Lines: Case of NGC 1097 2012 Wei Two Populations of Molecular Clouds in the Antennae Galaxies 8 2012 Sliwa Luminous Infrared Galaxies with the Submillimeter Array. III. The Dense Kiloparsec Molecular Concentrations of Arp 299 2012 Espada Disentangling the Circumnuclear Environs of Centaurus A: Gaseous Spiral Arms in a Giant 6 2013 Koenig The NGC 1614 . Molecular gas feeding a "ring of fire" 2013 Costagliola High-resolution mm and cm study of the obscured LIRG NGC 4418 . A compact obscured nucleus fed by in-falling gas? 4 2013 Sakamoto Submillimeter Interferometry of the Luminous Infrared Galaxy NGC 4418: A Hidden Hot Nucleus with an Inflow and an Outflow 2013 Hsi-An Formation of Dense Molecular Gas and at the Circumnuclear Starburst Ring in the Barred Galaxy NGC 7552 2 2013 Hwang Dust Properties of Local Dust-obscured Galaxies with the Submillimeter Array 2013 Sliwa Luminous Infrared Galaxies with the Submillimeter Array. IV. 12CO J = 6-5 Observations of VV 114 2013 Hirashita Properties of free-free, dust and CO emissions in the starbursts of blue compact dwarf galaxies 0 2004 2006 2008 2010 2012 2014 2013 Izumi Submillimeter ALMA Observations of the Dense Gas in the Low-Luminosity Type-1 Active Nucleus of NGC 1097 2014 Kuo Measuring Mass Accretion Rate onto the in M87 Using Faraday Rotation Measure with the Submillimeter Array SMA on Nearby Galaxies • SMA strengths: • sub-mm, sub-arcsec, wide-band, sensitivity, polarimetry, southern+northern sky, ... • good at warm/active sources, multi-line studies, ... • Spectroscopy • U/LIRG Legacy Project • U/LIRGs at highest resolutions • Feedback/Outflows • Warm/Dense Gas in Active Galactic Nuclei • Polarimetry

• ... SMA 10 @ Cambridge, MA, 2014 May Spectroscopy

SMA 10 @ Cambridge, MA, 2014 May New Extragalactic H2O Maser Humphreys et al. (2005) NGC 3079 (Seyfert 2) credit: credit: NASA, G. Cecil

H2O(61,6-52,3), Eu=643 K

H2O(31,3-22,0), Eu=205 K

Vsys New Extragalactic HNC Maser? Aalto et al. (2009) 12.2 S. Aalto et al.: HNC in Arp 220 485 10 L⦿ Cartoon of western nucleus 3–2 line can become inverted. For a column density of Arp 220 double nuclei @ 2cm 15 2 24 − face−on view N(HNC) = 10 cm− (assuming N(H2) = 10 through ST ST 9 1 and X(HNC) = 10− ), a line-width of 80 km s− ,acontin- uum background model as the one described in Downes & Eckart (2007), the HNC 3–2 line would be a maser with CND TB = 34 K, τ = 0.16 and Tex = 58 K. This simulation V (ST) is done with RADEX− code modified− by us to include the mid-IR and IR transitions of HNC. The on-line version of RADEX (van der Tak et al. 2007)doescurrentlynotinclude 1", 360 pc Norris (1988) these IR transitions of HNC. This is to illustrate that HNC S. Aalto et al.: HNC in Arp 220 483 maser action can be achieved with the approximate bright- Amplification of ness, excitation and optical depth as indicated in the simple background continuum PLot file version 5 created 31-MAY-2007 17:10:25 Plot file version 203 created 25-JAN-2008 17:04:15 BOTH: ARP220 IPOL HNC IMLI.MOM0.2 ARP220 RA 15 34 57.22123 DEC 23 30 11.5210 HNC IMLI.CLN.1 through ST model in the previous section. We furthermore note that in 20 40 60 this model, the HNC 3–2 line is by far the most luminous of 300 21.5 μm the rotational lines. The 1–0 and 2–1 lines are also masering, HNC 3−2 36 K but with lower brightness temperatures. The J = 4–3 line is 23 30 13.0 250 West nucleus not masering and is fainter by a factor of 6. b) Radiative pumping:bothHCNandHNChavedegener- 200 12.5 ate bending modes in the IR through which they can be fobs=267 GHz pumped (see Sect. 1). HNC is pumped via 21.5 µmcontin- 150 uum and the pumping of HNC may start to become effective 12.0 when the IR background reaches a brightness temperature of 100 T 50 K at that wavelength. The inner 30 pc CND would B ≈ MilliJY/BEAM then serve as a source of mid-IR photons to pump the HNC 11.5 50 molecules in the surrounding torus. According to Downes &

DECLINATION (J2000) Eckart (2007), the brightness temperature at 1.3 mm is 90 K. 0 Sakamoto et al. (2008)reportthe860µmbrightnesstemper- 11.0 ature of the west nucleus to be (90–160) K, where the range -50 corresponds to the range of possible source structure (i.e., aGaussianoradisc).Thenuclearmid-IRtemperaturemay 10.5 -100 therefore be as high as 160 K, although the mid-IR emission 15 34 57.35 57.30 57.25 57.20 57.15 RIGHT ASCENSION (J2000) 4.5 5.0 5.5 6.0 6.5 with brightnesstemperatureof 160 K for the dustin the CND Grey scale flux range= 10.00 72.57 Kilo JY/B*M/S 4500 5000Mega 5500 FELO-LSR 6000 Cont peak flux = 7.2568E+04 JY/B*M/S Average over area in X: 132 132 in Y: 130 130 Fig. 6. The figure shows a fit of a simple model (see text) to the cen- likely has a solid angle that is less than 2π as seen by the Levs = 7.257E+03 * (2, 4, 6, 8, 10) Freq smoothing parms: 1. 3.00 5.0V(opt) tral spectrum of the western nucleus, and a face-on cartoon of the ST ST, it is luminous enough to affect significantly the excita- and CND. The fit includes: 1) emission from the surrounding torus: τ τ τ tion of the ST – at least its inner parts. Furthermore, mid-IR Fig. 2. HNC J = 3–2 spectrum of the western nucleus of Arp 220. The T = T (1 e− ); 2) maser emission: T = T e− + T (1 e− ), where x − b x − imaging by Soifer et al. (1999)showedthatthebrightness spectrum has been Hanning smoothed resulting in a reduced intensity in Tb is (the enhanced) background temperature; 3) emission from the PLot file version 3 created 21-MAR-2007 14:40:00 τ temperature is around 100 K in the 100 pc region around BOTH: ARP220 IPOL HNC CELL.MOM1.1 the narrow component. The spatial resolution is 0$$. 48 0$$. 32. The broad CND: T = T0e− ,whereT0 is the brightness temperature from the ≈ 5200 5400 5600 the western nucleus. The mid-IR photons for the broad com- and narrow components are marked with a dashed curve.× inner disc (from a RADEX-model, with collisions only and modelled as a Gaussian, emission damped by foreground cloud). There are three ponent may therefore be from the 100 pc western disc it- curves: Fit1: τ = 0.15, T = 12, Fit2: τ = 0.08; T = 30, and Fit3: self. Modelling the geometry, radiation field, gas density and − 0 − 0 23 30 12.8 PLot file version 36 created 05-DEC-2007 16:12:52 τ = 0.04; T0 = 40. temperature will help determine what fraction of the HNC- CONT: ARP220 5162.3 KM/S IPOL ARP220 IMLIN.CLN.1 − 12.6 emission from the ST that could be affected by radiative 23 30 12.0 5162.3 KM/S 5191.7 KM/S 5221.1 KM/S 5250.5 KM/S overall line shape, broadening it. Thus, for the given conditions pumping. 12.4 11.5 above, we find a range of maser optical depths of τ = 0.04 − 12.2 11.0 to τ = 0.25 where the higher absolute optical depth is for a 12.0 − 3.3. Physical conditions in the ST 23 30 12.0 5279.9 KM/S 5309.4 KM/S 5338.8 KM/S 5368.2 KM/S small contribution from the CND. Note, however, that the dusty 11.8 graybody of the CND may prevent much line emission emerging 11.5 The broader part of the nuclear spectrum has a line width of 500– from the CND depending on dust optical depth, dust clumpiness 1 11.6 550 km s− and a peak flux of 100–150 mJy (11–18 K) (Fig. 2). 11.0 and structure.

DECLINATION (J2000) 11.4 We identify this feature with the surrounding, 90 pc torus (ST) 23 30 12.0 5397.7 KM/S 5427.1 KM/S 5456.5 KM/S 5486.0 KM/S The shape of the observed spectrum can be reproduced 11.2 for which Downes & Eckart (2007)findaCO2–1brightness 11.5 within the noise level. Testing a more sophisticated model re- temperature of 40 K and Sakamoto et al. (2008)findaslightly 11.0 11.0 quires higher signal-to-noise observations at higher resolution, higher CO 3–2 brightness of 50 K. Thus, roughly, the CO 2– 10.8 where the ST and CND contributions can be resolved. Below we

DECLINATION (J2000) DECLINATION 5515.5 KM/S 5544.9 KM/S 5574.4 KM/S 5603.9 KM/S 1/HNC 3–2 brightness temperature ratio is only 2–4 for the ST 23 30 12.0 discuss possible mechanisms for HNC population inversion. 15 34 57.30 57.25 57.20 57.15 outside of the luminous, narrow, maser-like feature. RIGHT ASCENSION (J2000) 11.5 Grey scale flux range= 5200.0 5720.0 Kilo M/S Such a low line ratio is consistent with both possible pump- Cont peak flux = 5.6999E+06 M/S 11.0 Levs = 5.200E+06 * (1.010, 1.020, 1.030, 1.040, ing scenarios discussed above. For the physical conditions sug- 1.050, 1.060, 1.070, 1.080, 1.100) 3.2. Conditions for population inversion 23 30 12.0 5633.4 KM/S 5662.8 KM/S gested by Downes & Eckart (2007)fortheSTitisdifficult to 11.5 In the previous section we propose that population inversion can 4 3 Fig. 1. Top panel:integratedHNC3–2lineemission(naturally get such a low CO/HNC line ratio. At n = 10 cm− ,thedensity 1 1 11.0 take place between the rotational levels 3 and 2 and we show is too low to effectively excite the HNC 3–2 line to the required weighted). The grey scale ranges from 10 to 73 Jy beam− km s− . 1 1 the line profile in a simple case of amplification by the HNC level unless the HNC abundances are extreme (see Sect. 3.6 for Contour levels are 7.3 Jy beam− km s− (2, 4, 6, 8). Crosses mark the 15 34 57.30 57.25 57.20 RIGHT ASCENSION (J2000) molecule. It is possible to either collisionally or radiatively pump adiscussionofHNCabundances).However,iftheSTisvery position of the 277 GHz continuum peaks.× The synthesized beam is in- Cont peak flux = 3.3196E-01 JY/BEAM Levs = 4.900E-02 * (2, 3, 4, 5, 6) HNC: 6 3 dicated in the box in the lower left corner. Lower panel:theHNC3–2 clumpy and clump densities can approach n = 10 cm− then 1 a) Collisional pumping:atgasdensitiesapproachingn = the observed line ratio can be understood. There is a large num- velocity field. The grey scale ranges from 5200 to 5720 kms− and the 6 3 1 1 Fig. 3. Naturally weighted channel maps, not smoothed in velocity. 10 cm− and kinetic temperatures around 150 K the HNC ber of SNRs in this area so it is likely that there is compressed, contours range from 5252 to 5616 km s− with steps of 52 km s− . Crosses mark the radio continuum positions of the two nuclei. Contour levels are 49 mJy (2, 3, 4, 5, 6). Peak flux is (330 50) mJy/beam. × ± 3. Discussion The HNC 3–2 spectrum of the western nucleus shows a broad feature upon which a narrow, luminous feature (Vopt = self-absorption, implying that the CO-West disc has a tempera- 1 ture gradient increasing inwards. 5440 km s− )issuperposed(seeSect.2.2.1 and Fig. 2). The peak brightness temperature of the HNC 3–2 emission fea- We suggest that the luminous, narrow HNC 3–2 emission ture is 39 K in our 0$$. 4 beam. In the same velocity range, features could be caused by weak maser action in the cooler CO 2–1≈ and 3–2 (at similar resolution) instead show deep, part of the nuclear disc – amplifying instead of absorbing the 1 narrow (100 km s− )absorption(Downes & Eckart 2007; background continuum from the inner hot, dusty part of the disc. Sakamoto et al. 2008). The deepest CO 2–1 absorption occurs If the narrow, luminous feature is indeed emerging only in front 1 at 5450 km s− ,butother,lessprominent,absorptionfeatures of the 0$$. 19 continuum source, the undiluted brightness temper- show up at higher velocities. In CO 3–2 the absorption oc- ature is 156 K. 1 curs in the range 5400–5450 km s− (transferred from the VLSR Previous claims of weak masering in other molecules of Sakamoto et al. 2008). Downes & Eckart (2007)suggest (besides OH) include methanimine (Salter et al. 2008), and that this is due partly to absorption of hot continuum, and CO formaldehyde (Araya et al. 2004). 1070 l=1f 1070 v2 =1 J=4 Firstl=1e Extragalactic Rotational lines 1060

l=1f 1050 J=3 1050 l=1efrom Vibrationally Excited HCN 1040 J=2 Sakamoto et al. (2011)

1030 F (rest) [GHz] F (rest) [GHz] 1030 J=1 265 266 267 268 355 356 357 358

50 PQR0.5 NGC HCN(v2=1;J=3-2) 0.8 HCN(v2=1;J=4-3) Energy / k [K] v=0 J=4 40 4418 0.4 30 0.6 J=3 20 0.3 J=2 10 0.4 J=1 ±1 ±1 0 HJ=0C N 0.2 1070 Flux Density [Jy] l=1f Flux Density [Jy] v2 =1 J=4 0.2 0.1 l=1e

1060

HCN Energy [K] Energy HCN l=1f 1050 J=3 l=1e (4-3)

0.0 0.0 + (3-2)

1040 + + + J=2 2500 2000 (HCN) 2500 2000 (HCO ) 2000 1500 (HCN) 2500 2000 (HCO )

1030 263 264 265 266 352 353 354 355 J=1 HCO HCN(4-3) HCN(4-3)

F (obs.) [GHz] HCO F (obs.) [GHz] HCN(3-2) HCN(3-2)

50 PQR 50 11.1 Energy / k [K] v=0 J=4 NGC 4418 10 L⦿ 40 • Tvib ~ 230 K 3030 J=3 • IR-pumping (14μm) 20 J=2 • Second extragalactic ro-vib 1010 J=1 0 line @ sub/mm, after HC3N 0 J=0 60"=10 kpc SDSS A&A 556, A66 (2013)

More Vibrationally-Excited Molecules Arp 220 Martin et al. (2011)

CH3CN v8=1 (Tvib ~ 400-450 K), HC3N v7=1,2, v6=1 (Tvib ~ 400 K) (Eu/k ~ 700 K) (Eu/k ~ 450, 700, 850 K) Costagliola et al. (2013) S

NGC4418 ν (mJy/beam)

Sky Freq. (GHz) HNC v2=1 (Tvib ~ 350 K), HC3N v7=1,2, v6=1 (Tvib ~ 315 K) (Eu/k ~ 690 K) Fig. 5. Spectra extracted from the central beam of MERLIN and SMA observations in NGC 4418. The data are drawn as shaded histogram, the solid black line represents the total profile of our LTE best-fit model, and LTE emission from single molecular species is drawn in blue. Panels c) and d) show a comparison of HI and CO 2–1 profiles. Dashed vertical lines mark the velocity centroid of the two components of CO emission, SC 1 at 2090 and RC at 2180 km s− LSR (see Sect. 4.1.1). The redshifted Gaussian components (RC), associated with a possible in-flow, are drawn in red in both panels. The conversion between flux density and brightness temperature is discussed in Sect. 4.2.1.TheresultsoftheGaussianfitfor each line are shown in Table 3.

1 components to be separated far enough in velocity to be con- emission at 2180 km s− (RC). The two components have similar sidered independent as far as radiative transfer is concerned. We sizes and orientation, but appear to be slightly displaced, with an therefore did not include radiative transfer in our model. This is offset of about 0"". 1intheSE-NWdirection.Thisoffset between discussed in more detail in Sect. 4.2.1. the two components is responsible for the observed velocity gra- The parameters that fit the data cube are reported in Table 4. dient at PA 135◦.Thissimplemodelcanreproduce90%ofthe The model is consistent with two main velocity components, observed CO emission, a comparison with the data is shown in 1 asystemiccomponentat2090kms− (SC) and redshifted Fig. 6.

A66, page 8 of 18 The Astrophysical Journal,743:94(19pp),2011December10 Rangwala et al.

Vibrationally-Excited(a) Molecules New tools to study hot molecular gas in luminous galactic nuclei - suitable for ALMA high-res imaging

Arp 220

(b) Herschel detection up to CO(13-12), Eu/k=503 K HCN(17-16), Eu/k=650 K at 1.5 THz

no high-res. imaging at 1.5 THz (c) for some time

Rangwala et al. 2011 Figure 1. Herschel SPIRE-FTS spectrum of Arp 220. The spectrum shows the FIR continuum between 190 and 670 µm in (a). Red solid points in (a) are the continuum measurements from the SPIRE photometer and the dotted curves show the photometer bandpasses with arbitrary normalization. Line identifications are shown for several molecular and atomic species in (b) and (c) with like colors for like species. (A color version of this figure is available in the online journal.)

FTS spectrum we start resolving the broad spectral lines towards rest frequencies, and integrated fluxes. The FWHMs mentioned the higher frequency end. We use a sinc function of variable in this section are from our sinc line fitting. FWHM to fit the emission and absorption lines using the MPFIT WedetectaluminousCOemissionladderfrom J 4–3 to J package in IDL. In Table 1 we report the integrated line fluxes 13–12 and several water transitions with total water= luminosity= 1 in Jy km s− for emission lines and equivalent widths in µmfor comparable to CO. Compared to the velocity widths measured the absorption lines along with their 1σ statistical uncertainties. in CO from the ground-based observations (e.g., SC97; Greve The observed spectral line shape suffers from small asymmetries et al. 2009), the higher-J CO lines in our spectrum are much from instrumental effects, and the line shapes could be affected narrower. These high-J CO lines are not resolved and follows by blending from weak lines. The effect on the line fluxes from the resolution of FTS, which at the frequency of J 12–11 line is 1 = asymmetries is expected to be !2%. 310 km s− .Thissuggeststhateitherthehigh-J CO emission is∼ coming from only one of the nuclei or that it is a dynamically 3. LINE IDENTIFICATIONS separate component from the low-J lines. On the other hand, the water lines are much broader with a velocity FWHM of 1 The spectrum (Figure 1)showstheFIRcontinuumandthe 500 km s− implying that this emission could be coming from detection of several key molecular and atomic species and ∼both nuclei. The observed velocity separation between the two 1 Figure 2 shows zoom-in views of some of the spectral lines nuclei in Arp 220 varies considerably from 50 km s− (Downes 1 discussed in this section. In Table 1, we list their transitions, &Solomon1998;Sakamotoetal.2008)to250kms− (Scoville

3 U/LIRG Legacy Project

SMA 10 @ Cambridge, MA, 2014 May SMA Legacy Survey of U/LIRGs Ultra-Luminous InfraRed Galaxies Wilson et al. (2008) Iono et al. (2009) Silwa et al. (2012,13)

DL < 200 Mpc log LFIR > 11.4 dec. > -20°

Observed 14 (out of 39).

CO(3-2), CO(2-1), HCO+ (4-3), continuum

~1kpc (1") resolution red circle=F.o.V

log LFIR SMA Legacy Survey of U/LIRGs Wilson et al. (2008) Iono et al. (2009) CO(3-2) res. ≈ 1kpc

CO(3-2) is detected as ~1 kpc peaks. ~50% of total is in these peaks. 9 M(H2)~10 M⊙ (Wilson et al. 2008) SMA U/LIRG Survey : Correlations LFIR - ΣH2(peak) Tdust - ΣH2(peak) Wilson et al. (2008) ] -2 pc ⊙ (peak) [M (peak) H2 Σ

log log >99% correlation >99% correlation prob. of false correlation = 0.0014 prob. of false correlation = 0.0016

log LFIR [L⊙] Tdust [K]

LFIR/MH2(peak) - ΣH2(peak) ] ⊙ More gas to the center (~kpc) /M ⊙ ⇒ More Luminous (higher LFIR)

w/o increasing efficiency (Mgas/LIR) (peak) [L (peak) 2

H c.p. CO(1-0) studies by

/M Scoville et al. 1991 (ApJ);

FIR Okumura et al. 1991 (IAUS 146) no correlation

log L log ⇒ Warmer ISM (higher Tdust) -2 log ΣH2(peak) [M⊙pc ] SMA U/LIRG Survey : Correlations Iono et al. (2009)

CO(1-0) - LFIR ] 2 ] ⊙ pc [L -1 FIR L

SMA data et al. (2009) Chung

[K km s non-linear

L'CO(1-0) [K km s-1 pc2] CO(3-2)

L' 0.93 L'CO(3-2) LFIR HCN(1-0) - LFIR near linear

LFIR [L] LCO(3-2) - LFIR : near linear correlation Solomon et al. (1992) et al. (1992) Solomon

slope=0.97 (2004) Gao & Solomon • CO(3-2) traces SF dense mol. gas linear • LFIR/LCO(3-2) = (dense gas) SFE ~ constant U/LIRG: Highest resolution

SMA 10 @ Cambridge, MA, 2014 May Arp 220 @ 860μm, 0.23" Sakamoto et al. (2008)

0 20 40 60 80 100

(a) Arp 220 Arp 220 system Arp 220 W: CO(3-2) 1 nuclear disks Deconvolved Size d ~ 0.15"-0.22" = 50-80 pc 0

Arp 220W Arp 220E Deconvolved (peak) Tb Dec. offset [arcsec] -1 Tb = 90-160 K ≤ Tdust (τ860 ~1) outer disk 0.38" x 0.28" 300 pc Luminosity -2 (Sakamoto et al. '99) 1 0 -1 -2 R.A. offset [arcsec] 4 2 11 5100 5200 5300 5400 5500 5600 0 50 100 150 200 Lbol ≈ σTd × π d ≥ (2-3)×10 L

(b) Arp 220 (d) Arp 220 CO(3-2) Luminosity surface density 860μm continuum0.86 mm 1 1 7.6 -2 Σ(Lbol) ≥ 10 L pc

0 0

Dynamical mass (W disk ≈ edge-on) 8 Dec. offset [arcsec] Dec. offset [arcsec] Mdyn (r ≦ 40pc) ~ 6x10 M -1 -1 Luminosity-to-Mass ratio 0.38" x 0.28" 300 pc 0.25" x 0.21" 300 pc -2 -2 1 0 -1 -2 1 0 -1 -2 L/M ≳ 400 L /M (r≦ 40pc) R.A. offset [arcsec] R.A. offset [arcsec]

NGC 4418 @ 1300/860/450μm, Sakamoto et al. (2013)

0.4/0.3/0.2" Costagliola0 5 et al.10 (2013)15

0 2000 4000 6000 8000 (k) 860 µm 1 NGC 4418 3 (a) CO(3−2) mom0

2 0

1 Dec. offset [arcsec] −1 2000 2050 2100 2150 2200

0.35" x 0.26" 50 pc 0 (f)1 0 HCN(4− 13) 1 R.A. offset [arcsec]mom1 Dec. offset [arcsec] −1 11.1 LIR=10 L⊙ 9.8 LB =10 L⊙ −2 34 Mpc 0 5.5 kpc 0.69" x 0.55" 300 pc −3

3 2 1 0 − 1 − 2 − 3 Dec. offset [arcsec] 0.25 R.A. offset [arcsec] −1 0 10 20 30 0.36" x 0.26" 50 pc (a) NGC 4418 continuum, 850 µm 0.20 (l)1 0 450 µm− 1 a ~20 pc dusty core 1 R.A. offset [arcsec] 11 0.15 Lbol ~ 10 L⊙ ~ Lgalaxy

0 0.10 Lbol/Mdyn~500 L⊙/M⊙ Amplitude [Jy] d ≤ 0.1" (20pc) 25 -2 0.05 NH≥10 cm (Compton thick) Dec. offset [arcsec] error bar=±1 Tb ~ 160 K −1 0.23" x 0.15" 50 pc 0.00 0 100 200 300 400 500 600 1 0 − 1 UV length [k] R.A. offset [arcsec] Submm-diagnosis of Lbol-source

105

LLEddingtonEdd/M / BHM

104

LSB/MYoung ] ★ young SB (<10Myr) • O

3 or / M 10 • O

Lbol/Mdyn, Arp 220, NGC 4418 AGN+SB L / M [L 102

Starburst99+KroupaStarburst IMF

10-100 M O• 1 10 1-100 M O• SMA obs. 0.1-100 M O• Arp 220: Sakamoto et al. 2008 N4418: Sakamoto et al. 2013 1 10 Time [Myr] Feedback/Outflows

NGC 1266 → Next Talk

SMA 10 @ Cambridge, MA, 2014 May Feedback: Mol. Superbubbles Sakamoto et al. (2006, 2011) 0 1000 2000 3000 4000 5000 6000

Southern Starburst Galaxy (a) NGC 253 12CO(2-1) NGC 253 10

5

0

-5 Dec. offset [arcsec]

-10 2' = 2 kpc 24μm/8μm/FUV 1.1" x 1.1" 100 pc ?

(Sakamoto, Mao, Matsushita et al. 2011) (Sakamoto, Ho, Iono, Keto, Mao et al. 2006) Cavities in the central molecular zone 0 500 1000 1500 2000

30 NGC 253 12CO(J=2-1) Feedback: Mol. 20 Superbubbles

10 Cavity Superbubble ? If so,

• D ~ 100 pc • Age ~ 1 Myr 0 46 • dV ~ 100 km/s • E ~ 10 J ~ 100 ESN -10 Dec. offset [arcsec]

0 5 10 0 5 10 0 5 10 0 5 10 0 5 10 -20 20 120.0 km/s 125.0 km/s 130.0 km/s 135.0 km/s 140.0 km/s 15 1.3" x 1.3" 100 pc 10 -30 5 45 30 15 0 -15 -30 -45 0 R.A. offset [arcsec] -5 20 145.0 km/s 150.0 km/s 155.0 km/s 160.0 km/s 165.0 km/s 15 10 5 ARC SEC 0 -5 20 170.0 km/s 175.0 km/s 180.0 km/s 185.0 km/s 190.0 km/s 15 10 5 0 -5 30 20 10 30 20 10 30 20 10 Sakamoto et al. 2006 ARC SEC 0 500 1000 1500 2000

30 NGC 253 12CO(J=2-1) Feedback: Mol. 20 Superbubbles

10 Cavity Superbubble ? If so,

• D ~ 100 pc • Age ~ 1 Myr 0 46 • dV ~ 100 km/s • E ~ 10 J ~ 100 ESN -10 0 2 4 6 8 0 2 4 6 8 0 2 4 6 Dec. offset [arcsec] 8 0 2 4 6 8 0 2 4 6 8

5 330.0 km/s 335.0 km/s 340.0 km/s 345.0 km/s 350.0 km/s 0 -20 -5 -10 1.3" x 1.3" 100 pc -15 -30 -20 5 355.0 km/s 360.0 km/s 365.0 km/s 45 370.0 km/s 30 375.015 km/s 0 -15 -30 -45 R.A. offset [arcsec] 0 -5 -10 -15 -20 5 380.0 km/s 385.0 km/s 390.0 km/s 395.0 km/s 400.0 km/s 0 -5 -10 ARC SEC -15 -20 5 405.0 km/s 410.0 km/s 415.0 km/s 420.0 km/s 425.0 km/s 0 -5 -10 -15 -20 10 0 -10 -20 10 0 -10 -20 10 0 -10 -20 ARC SEC Sakamoto et al. 2006 Feedback: Mol. Outflow

ALMA CO(1-0)

SMA cavity

(Bolatto(Garcia-Burillo et al. 2013) et al. 2000)

-1 dMw/dt ≈ 9 M⊙ yr ≥ SFR Galactic Molecular Outflow from CO line wings Sakamoto, Ho, Peck (2006)

11 NGC 3256, merger, LIR=4x10 L⦿

0 20 40 60 80 100 120

00 (a) NGC 3256

12CO(J=2-1) 05

10

15

20 Declination (J2000)

25

30 2.0" x 1.0" 1 kpc -43° 54 35 10 h 27 m 52.5 s52.0 s 51.5 s 51.0 s 50.5 s 50.0 s 49.5 s Right Ascention (J2000) Galactic Molecular Outflow

0 20 40 60 80 100 120

from CO line wings 00 (a) Sakamoto, NGCHo, 3256 Peck (2006) 12CO(J=2-1) 05

10

15

20 Declination (J2000)

25

1 kpc 30 0 2 4 6 [K] 2.0" x 1.0" 1 kpc -43° 54 35 NGC 3256 h m s s s s s s s 3100 10 SMA 27 52.5CO(2 52.0 − 1) 51.5 51.0 50.5 50.0 49.5 Right Ascention (J2000)

3000

2900

2800

Velocity [km/s] 2700 • High-velocity gas (molecular outflow) 2600 outflow rate ~ 10 M/yr ~ SFR 1 kpc • First one found from CO line wings −20 −10 0 10 20 offset alongoffset P.A.= [arcsec]−90 deg [arcsec] Double-Outflow System

CO(1−0) [−223,−751] km/s Sakamoto et al. (2014) 10 NGC 3256 [+224,+752] km/s ALMA

CO(3-2)(b) −199 km/s −260 km/s −321 km/s −381 km/s −442 km/s 6 +200 km/s +260 km/s +321 km/s +382 km/s +442 km/s 5 4

2

0 0

−2 Dec. offset [arcsec] −4 Dec. offset [arcsec] −5 −6 0.6" x 0.4" 300 pc 4 3 2 1 0 − 1− 2− 3− 4 3 2 1 0 − 1− 2− 3 3 2 1 0 − 1− 2− 3 3 2 1 0 − 1− 2− 3 3 2 1 0 − 1− 2− 3 R.A. offset [arcsec] R.A. offset [arcsec] R.A. offset [arcsec] R.A. offset [arcsec] R.A. offset [arcsec] −10 2.9" x 2.6" 500 pc 10 5 0 − 5 −10 R.A. offset [arcsec] • Two molecular bipolar outflows • from N nucleus: pole-on, wide opening angle • from S nucleus: edge-on, collimated bipolar jet • total dM/dt ~ 100 M/yr Galactic Molecular Outflows from P-Cygni Line Profiles Sakamoto et al. (2009) 12.2 10 L⦿ (a) Arp 220 (b) Arp 220 + 1 HCO+(4-3)HCO+(4-3) HCO+(3-2)HCO (3-2)

0

Dec. offset [arcsec] -1 0.33" x 0.28" 300 pc 0.45" x 0.34" 300 pc Arp 220 system (c) Arp 220 (d) Arp 220 1 CO(3-2)CO(3-2) 0.87 mm

nuclear disks

0 Arp 220 double nuclei @ 2cm

Dec. offset [arcsec] -1 0.34" x 0.28" 0.34" x 0.28" 300 pc 1 0 -1 1 0 -1 Arp 220W R.A. offset [arcsec] R.A. offset [arcsec] Arp 220E 1", 360 pc Norris (1988) Sakamoto et al. (1999)

outer disk Galactic Molecular Outflows from P-Cygni Line Profiles Sakamoto et al. (2009) Arp 220E Arp 220W

200 Arp 220+ E 200 Arp 220+ W HCOHCO+(4-3)(4-3) HCOHCO+(4-3)(4-3) (a) Arp 220 (b) Arp 220 100 100 + 1 HCO+(4-3)HCO+(4-3) HCO+(3-2)HCO (3-2) 0 0.0 0 0.0 -100 -0.5-100 4500 4700 4900 4500 4700 4900 1 HCN(4-3) Velocity 1 HCN(4-3) Velocity 0 -1.0 -200 -200 -0.5

] ]

-1 + -1 + 100 HCOHCO+(3-2)(3-2) 100 HCOHCO+(3-2)(3-2) Dec. offset [arcsec] -1 0.45" x 0.34" 300 pc 0.33" x 0.28" 300 pc Arp 2200 system 0.0 0 0.0 1 -0.5 1 -0.2 -1.0 (c) Arp 220 (d) Arp 220 1 CO(3-2)CO(3-2) 0.87 mm CO(3-2) CO(3-2) 400 CO(3-2) 400 CO(3-2)

Intensity [mJy beam CO(3-2), 0.5" data Intensity [mJy beam CO(3-2), 0.5" data Vsys (W)

0 200 Vsys (E) 200 nuclear disks Dec. offset [arcsec] -1 0 0.0 0 0.0 0.34" x 0.28" 0.34" x 0.28" 300 pc -0.5 -0.2 1 1 -1.0 1 0 -1 1 0 -1 -0.4 R.A. offset [arcsec] R.A. offset [arcsec] 4500 5000 5500 6000 6500 4500 5000 5500 6000 6500 Velocity (radio, LSR) [km s-1] Velocity (radio, LSR) [km s-1]

Blueshifted + Redshifted absorption emission = P-Cygni profile Arp 220W ⇒ nuclear outflows/winds Arp 220E Arp 220E

outer disk Galactic Molecular Outflow from Orthogonal Vel. Gradients Aalto et al. (2012)

Plot file version 31 created 29-OCT-2010 16:34:35 NGC1377 RA 03 36 39.11282 DEC -20 54 07.8000 N1377 LOW.INV.1 1.6600

500

CO wings 400

300 MilliJy/beam

200Jy/beam

100

0 0 1600 1650 1700 1750 1800 1850 1900 1600 1700Kilo VELOCITY 1800 1900 Average over area in X: 240 267 in Y: 233 272 V [km/s] Freq smoothing parms: 3. 3.00 3.0 km/s orthogonal Vel. gradients Plot file version 138 created 08-MAY-2011 20:54:11 NGC1377 RA 03 36 39.07714 DEC -20 54 07.0000 N1377 CO.INV.1

dust cone 0.2200

150

100 MilliJy/beam Jy/beam

50

00

1600 1650 1700 1750 1800 1850 1900 1600 1700Kilo VELOCITY 1800 1900 Average over area in X: 512 512 in Y: 513 513 Freq smoothing parms: 2. 2.00 7.0 km/s

-1 30" = 3kpc Vout = 140 km/s dM/dt > 8 Myr Roussel et al. (2006) age ~ 1.4 Myr Inflow+Outflow System

1.8 2.0Sakamoto2.2 et al.2.4 (2013)2.6 2.8

(i'+z')/(g'+r') (i’+z’)/(g’+r’) 20 NGC 4418 10 1

0

0.9

Dec. offset [arcsec] −10

−20

1.5" 1 kpc V-VSYS [km/s] 30 20 10 0 −10 −20 −30 R.A. offset [arcsec] reddening cone redshift absorption // semi-minor axis → inflow → outflow Gonzalez-Alfonso et al. (2012) Warm/dense Gas in Active Galactic Nuclei

SMA 10 @ Cambridge, MA, 2014 May Warm Gas in (active) Galaxy Nuclei

M51 L56 MATSUSHITA ET AL. Vol. 616 CO(3-2), SMA

The Astrophysical Journal,736:129(17pp),2011August1 Hsieh et al. where r is the radius of the clump derived in Section 3.2 and σrms is the three-dimensional intrinsic root-mean-square velocity dispersion, which is equal to √(3/8ln2)σint.Sinceweobserve 2 No. 1] Enhanced HCN Emission in NGC 1097 σint in one dimension, we need to multiply the observed σintL3by 3. The radius is adopted with the size of the clumps assuming that σint is isotropic. The results are shown in Table 3.Notethat the virial mass is for GMAs not GMCs, and we assume that the LETTER LETTER LETTER LETTER LETTER LETTER GMAs are bound structures. The ratio of the virial mass Mvir to LETTER LETTER LETTER Fig. 2.—Integrated intensityMgas map ofis the shown12CO (J p 3–2 in ) line. Table The gray3 scale.Wefoundthatthenarrowlineclumps and the synthesized beam are shown at the top and the bottom left corners, Sakamoto et al 1999 respectively. The plus signhave and the circleM arevir the/M samegas as inthat Fig. 1. are The contour more or less about unity, but are larger levels are 2, 3, 4, 6, 9, 12, and 15 j,where1j p 5.89 Jy beamϪ1 km sϪ1 (p9.63 K km sϪ1). Primarythan beam correction unity is for not applied. the broad line and dust lane clumps. Fig. 1.—Channel maps of the 12CO (J p 3–2 ) line of the central regionMatsushita of et al. (2004) M51. The synthesized beam (3Љ.9 # 1Љ.6 or 160 # 65 pc) is shown at the bottom We plot the general properties of the molecular gas mass of left corner of the first channel map. LSR velocities are shown at the bottom The compact nuclear emission also appears at redshifted Seyfert nucleus brighter in higher J COϪ1 individual12 molecular clumps in Figure 7.InFigure7(a), the right corner of each map. The systemic velocity is 472 km s (Scoville & velocities in interferometric CO (J p 1–0 ) observations by Young 1983; Tully 1974). The plus sign and the 36Љ (1.5 kpc) diameter circle Aalto et al. (1999) andhistogram Sakamoto et ofal. (1999). the gas Similarly, mass in- integrated over their size seems to be in each map represent the galactic nucleus determined from the 6 cm radio terferometric 12CO (J p 2–1 ) observations by Scoville et al. continuum peak (Ford et al. 1985) and the half-power width of the primary apowerlawwithasharpdropatthelow-massend.Thisisdue ⇒ warm/dense gas at thebeam, respectively. nucleus The contour levels are Ϫ6, Ϫ3, 3, 6, 9, 12, and 15 j, (1998) also show predominantly redshifted emission concen- where 1 j p 70.4 mJy beamϪ1 (p0.115 K). Primary beam correction is not trated 1Љ west of the nucleusto the but, sensitivity in addition,relatively limit, since weak the corresponding 3σ mass limit is The Astrophysical Journal,747:90(10pp),2012March10 applied. blueshifted emission east of the nucleus6 (see their Figs.Hsieh 1 and et al. 2). The integrated intensity27 of their10 blueshiftedM for emission a clump is only with a diameter of 3"".3(theaverage package AIPS, and the resultant angular resolution was 18% of that of the redshifted∼of the× clumpsemission, with$ in aTable deconvolved3). In Figure 7(b), the FWHM intrinsic NGC1097 CO(1-0), NMA 3Љ.9 # 1Љ.6( 160 # 65 pc) at a position angle (P.A.) of 146Њ size for the blueshifted emission of less than 0Љ.5. In our map, withCO(2-1), natural weighting. We SMA made channel maps with 42 channel the redshiftedCO(3-2), emissionline wasSMA widths detected at have a peak-integrated a weak correlation in- with gas mass in the ring. In binning, which corresponds to a velocity resolution of 29.6 km tensity of 15 j at a beam size of3Љ.9 # 1Љ.6 . Assuming that the sϪ1. The typical rms noise level of the channel maps was redshifted and blueshiftedFigure emission8(a), inJ wep 3–2 show has the same azimuthal variation of ΣH2 calculated at Ϫ1 70.4 mJy beam , which corresponds toTsys, DSB 700 K. The intensity ratio as thatthe inJ emissionp 2–1 , the blueshifted peaks of emission the clumps is integrated over one synthesized half-power width of the primary beam at 345∼ GHz is 36Љ expected to have a signal-to-noise ratio of less than 2 j.We (1.5 kpc). We did not detect any continuum emission at the rule out the possibilitybeam. that the If systemic we assume velocity thatof the NGC cir- 1097 has a trailing spiral, then the Ϫ1 rms noise level attained of 18 mJy beam . cumnuclear moleculardirection gas differs from of that rotation of the host is galaxy clockwise from east (0◦) to north (90◦).

3. RESULTS The ΣH2 in the orbit-crowding region is roughly from 0◦ to 45◦, Channel and integrated-intensity maps in 12CO (J p 3–2 ) as and from 180◦ to 225◦. Note that the dust lane clumps typically shown in Figures 1 and 2 show a strong peak at the nucleus have ΣH2 similar to the narrow line ring clumps, which are lower at redshifted velocities (with respect to the systemic velocity of 472 km sϪ1; Scoville & Young 1983; Tully 1974). The than the broad line ring clumps in the orbit-crowding region. highest velocity of the compact nuclear emission is 616 km The average of the narrow line ring and dust lane clumps Ϫ1 ∼ ΣH2 s (as can also be seen in the position-velocity [PV] diagram 2 2 of Fig. 3), comparable to the highest velocity seen in the red- is 1800 M pc− , and 2300 M pc− for the broad line ring shifted wing of the 12CO (J p 3–2 ) spectrum taken with the clumps.∼ In Figure$ 8(b), the∼ velocity$ dispersion of clumps located James Clerk Maxwell Telescope (Matsushita et al. 1999). Our maps also show weak emission from the spiral arm to the in the orbit-crowding region and the dust lane is larger than that northwest of the nucleus (hereafter, the northwestern300 arm) pc at of the narrow line ring clumps. The average velocity dispersion blueshifted velocities. The spiral arm to the southeast (hereafter, Fig. 3.—PV diagrams (a)paralleland(b) perpendicular to the radio jet (P.A. 165Њ). The velocity resolution is 14.8 km sϪ1.Positionsaretheoffsets 1 1 the southeastern arm) was not detected. The previously pub- ∼ of the narrow line clumps is 50 km s− ,and 90 km s− 12 from the nucleus. The contour levels are Ϫ2, 2, 4, 6, 8, 10, and 12 j,where lished single-dish CO (J p 3–2 ) maps (Wielebinski et al. Ϫ1 Ϫ1 ∼ ∼ 1 j p 76.5 mJy beam forfor panel thea andbroad 47.8 mJy beam linefor ring panel b/.Dotteddust lane clumps. Given the similar line Kohno et al 2003 1999) also show the intensity asymmetry, i.e., stronger emis- horizontal lines show the systemic velocity. Dashed lines show the results of sion from the northwestern than from the southeastern arm. intensity-weighted linear fitting.brightnesses (Table 2)ofthepeaks,theincreased values Hsieh et al. (2008, 2011, 2012) ΣH2 12 1 1 + Figure 1. Left: HCN(J 3–2) integrated intensity map,Figure the 4. contoursTop image levels is the areCO( 2, 4,J 6,2–1)-integrated 8, 10, 12, 14σ map,where1 (contours)σ is overlaid 1.2 Jy beam−arekm probably s− . Middle: the results HCO ( ofJ increased3–2) line widths as indicated in = on the archival HST I-band (Filter1 F814= W)1 image (grayscale).12 Astrometry of = integrated intensity map, the contours levels are 2, 4, 6σ ,where1σ is 1.2 Jy beam− km s− .Right: CO(J 3–2) integrated intensityEquation map, (2). the On contours the other levels hand, are since the NH is proportional to 1 the1 HST image was corrected using background stars with known= positions. 2 2, 4, 8, 16, 32, 64σ ,where1σ is 4.5 Jy beam− km s− . The synthesized beam of the three images is 4"". 4 2"". 7(P.A. 14◦.6). the number density of gas and the line-of-sight path (i.e., optical Downloaded from The contour levels for 12CO(J 2–1) are 2σ,3σ,5σ , ...,20×σ,25σ ,and30= −σ 1 1= (A color version of this figure is available in the online(1σ journal.)2.3 Jy km s beam ). The IDs for the individual peaks of clumps depth), the higher ΣH2 of the broad line clumps may be due to = − − are marked. The CO-synthesized beam (1"". 5 1"". 0) is shown in the lower right either a larger number density, or a larger line-of-sight path. We corner. The bottom image is the HST NICMOS× Paα line image (color) overlaid will discuss these effects in Section 4. on the 12CO(J 2–1) contour. The contour levels are the same as in the upper Here, we compare our first interferometric HCN(J =3–2) and comparisons of both lines, we convolved the maps to 4"".4 2"".7 + image. = ×1 HCO (J 3–2) maps of NGC 1097, at an angular resolution of (P.A. 14◦.6). The 1σ noise levels are 11 mJy beam− for = = − 3.3. Young Star Clusters 4"", with the emission from the lower-J transitions. both lines. ∼ 6 7 (5.8 2.5) 10 M and (1.1 0.4) 10 M for 20 K and To check how the star formation properties are associated ± × $ ± × $ 13 2. OBSERVATIONS AND DATA REDUCTION50 K, respectively. With an assumption of constant CO3./H RESULTS2 with the molecular clumps in the ring, we compare the 6 cm abundance, if the Tex is !20 K, then the conversion factor we radio continuum, V-band, and Paα images with the molecular http://pasj.oxfordjournals.org/ We observed NGC 1097 with the Submillimeteradopted for the Array12CO line7 is overestimated by a factor of 3–5. 3.1. High Excitation∼ Dense Gasgas in images. the Circumnuclear The 6 cm radio continuum sources are selected (SMA; Ho et al. 2004). Two nights ofThe observations overestimation were is smaller (by a factor of 2–3) ifRegion the gas of NGCfrom 1097 the intensity peaks (Hummel et al. 1987)atanangu- is as warm as 50 K. We measured the 12CO(J 2–1) flux obtained in 2008 and 2010 with the compact configuration and = lar resolution of 2"".5. The V-band clusters (<13 mag) are se- the 345 GHz receivers. The uv-coverageshigher were than7.5–61.5 3σ to derive kλ the totalIn H Figure2 mass of1 we the nucleus show the and HCN(lectedJ 3–2), from the HCOHST+(JF555W3–2), image (Barth et al. 1995). These the ring.∼ The total flux of the nucleus12 and the ring are 490.2 V-band-selected= clusters= have a typical size of 2 pc (0.03), for both observations. The system temperatures (Tsys1 )were and CO(J 1 3–2) (Hsieh± et al. 2011) integrated intensity "" 59.3 Jy km s− and 2902.2 476.8 Jy km s−=,respectively.If and+ are suspected to be super star clusters by Barth et al. 150–300 K and 100–180 K, respectively,we adopt for intensity the tworatios of ±12CO(maps.J The2–1) HCN( and 12CO(J J 3–2)1–0) and HCO (J 3–2) maps show ∼ ∼ anuclearconcentrationandamolecularringsimilartothe= = = (1995). Pa=α clusters are identifiedJ in our HST F187N image sessions. of 1.9 0.2 and 1.3 0.2 for the nucleus and the ring (Paper I), described in Section 2.ThePaα clusters are identified by ± ± 1–0, J 2–1, and J 8 3–2 CO maps (Kohno et al. 2003=, The SMA correlator processes two intermediate-frequencythe MH2 of the nucleus and the ring are (1.6 0.3) 10 M SExtractor with main parameters of a detection threshold of 9 = ± × = $ (IF) sidebands separated by 10 GHz, withand2GHzbandwidth (1.4 0.3) 10 M ,respectively.Thus,thenucleusand2007;Hsiehetal.2008, 2011). The15σ ( nucleusDETECT_THRESH is the brightest)andaminimumnumberofpixelsabove ∼ ± × $ each in 2008. The bandwidth was upgradedthe to ring 4 GHz account for for each 10% and 90%feature of the inM allH2 thewithin maps, the 2 and kpc the relativelyathresholdof1( weakerDETECT_MINAREA molecular ). The number of deblend- IF after 2009. The upper/lower sidebandscircumnuclear were divided region, into respectively.ring coincides with the starburst ring.ingThe sub-thresholds ring consists is 50 of for giantDEBLEND_NTHRESH and 0.0005 for slightly overlapping 24 and 48 chunks of 104We calculate MHz width the virial in massmolecular of individual cloud clumps associations by (GMAs)DEBLEND_MINCONT at the scale of.Theparameterswerechosenbyawide GMAs of range of verifying and visual+ inspections. 2008 and 2010, respectively. In 2008, we observed the HCN(J 200–300 pc. However, since both of the HCN and HCO lines by guest on June 8, 2014 2rσ 2 We used 6 cm radio continuum and V-band-selected clus- = M haverms higher, critical density(5) than CO, the HCN and HCO+ GMAs 3–2) line (rest frequency: 265.886 GHz) in the lower sideband vir ters for the phenomenological comparison with the molecular with six antennas. In 2010, we observed both the HCN(J =areG expected to be the denser regions of the GMAs mapped in the + = 3–2) and the HCO (J 3–2; rest frequency: 267.558 GHz) previous CO lines. We identify a8 total of eight GMAs including lines simultaneously in= the lower sideband with eight antennas the one at the nucleus (Table 1). The GMAs are defined by the because of the availability of the 4 GHz bandwidth. intensity peaks with visual inspection. We show the spectra of We calibrated the SMA data with the MIR-IDL software the nucleus in Figure 2.Thespectrashowasymmetricprofiles, package. The bandpass calibrators are 3C84, Uranus and Ceres which are likely due to multiple underlying components. The for the first track, and 3C454.3 and 3C84 for the second track. blueshifted part is fainter than the redshifted part in all the lines. Callisto and Uranus were used for the absolute flux calibration The nucleus is brighter in HCN than HCO+ in both the low- for the first and second data set, respectively. The accuracy of and high-J lines. HCN(J 3–2) has better contrast between the the absolute flux calibration is 15% for both of the data. The nucleus and the starburst= ring than CO(J 3–2) and HCO+(J gain calibrators are the quasars∼ 0423–013 for the first track and 3–2). We demonstrate this by explicitly= comparing the con-= 0137–245 (phase)/0403–360 (amplitude) for the second track, tribution of the emission from the nucleus and the molecular respectively. Mapping and analysis were done with the MIRIAD ring in these lines. The HCN(J 3–2) integrated fluxes for = and the NRAO AIPS packages. We made the maps with natural the nucleus (radius: 0""–6"")andthering(radius:6""–13"")are 1 1 1 weighting, and with the data binned to 20 km s− velocity 34.4 3.4 Jy km s− and 89.9 6.7 Jy km s− ,respectively. resolution. The CLEAN processes were done in MIRIAD to Hence± the nucleus contributes 30%± 3% of the HCN(J 3–2) + ± = remove the sidelobes. The synthesized beams are 3"".7 2"".6 flux within 13"".TheHCO(J 3–2) integrated fluxes of the × = 1 (P.A. 10◦)fortheHCN(J 3–2) maps and 4"".4 2"".6 nucleus and the ring are 17.4 3.4 Jy km s− and 85.8 = − + = × 1 ± ± (P.A. 15◦)fortheHCO(J 3–2) maps. For the latter 6.7 Jy km s− ,respectively.Therefore,thenucleuscontributes + = − = 17% 3% of the HCO (J 3–2) flux within 13"".Theinte- Fig. 1. NMA maps of the CO (1–0) and HCN (1–0) lines, along with the optical image±12 of NGC 1097. The= central cross in figures (a), (b), and (c) marks 7 grated CO(h J m 3–2)s fluxes of the nucleus and ring are 1021 the peak of the 6 cm continuumThe SMA is(Hummel a joint project et between al. 1987). the Smithsonian The coordinates Astrophysical are α(J2000) = 02 46= 181 . 96 and δ(J2000) = 30◦16"28.""9. In each molecular Observatory and the Academia Sinica Institute of Astronomy and Astrophysics and 3778 Jy km s− . Hence the nucleus contributes− 21% 0.2% map, the NMA field ofand view is funded (60 by"" thefor Smithsonian CO and Institution 80"" for and HCN)the Academia is indicated Sinica. by a dashedof the total circle.12CO( AJ circumnuclear3–2) flux within starburst 13 .TheHCN( ringJ with± 3–2) a radius of r 10"" = "" = ∼ or 700 pc is shown by a red circle, and two red lines indicate the dust lanes along the bar. Attenuation due to primary beam pattern of each 10 m dish 2 has been corrected in these maps. (a) Integrated intensity map of CO. The synthesized beam is 7.""7 3.""9(540 270 pc) with a P.A. of 16◦.The 1 1 × 1 × − contour levels are 1.5, 3, 4.5, 6, 7.5, 9, 12, 15, and 18 σ, where 1σ =3.39Jy beam− kms− or 10.4Kkms− in Tb.Thiscorrespondstoaface-ongas 2 1 surface density, Σgas,of20.8 M pc− ,calculatedasΣgas =1.36 ΣH and ΣH =2.89 cos i (ICO/Kkms− ), where ICO is the velocity-integrated $ × 2 2 × CO intensity and i is the inclination of the disk (46◦ for NGC 1097; Ondrechen et al. 1989). (b) Integrated intensity map of HCN. The synthesized 1 1 beam is 10.""1 4.""4(710 310 pc) with a P.A. of 8◦.Thecontourlevelsare1.5,3,4.5,...,15,and16.5 σ,where1σ =0.711 Jy beam− km s− ×1 × − 1 1 or 3.54K km s− in Tb.(c)Intensity-weightedmeanvelocitymapofCO.Thecontourintervalis30kms− , and the systemic velocity of 1254km s− , determined by the fitting of this CO data, is indicated. Circular rotation seems to dominate the kinematics in the central r<10"" region (i.e., within the circumnuclear ring) in NGC 1097, whereas very strong non-circular motions along the dust lanes are suggested by isovelocity contours nearly parallel to the dust lanes. It is consistent with the claimed sheared gas flow along the bar, which is suggested from observations of magnetic field in NGC 1097 (Beck et al. 1999). (d) A V -band image of NGC 1097 (Quillen et al. 1995). The central 400 400 (28 kpc 28 kpc) region is displayed. A square "" × "" × indicates the area of the CO and HCN maps (60 60 ). "" × "" A&A 525, A18 (2011)

Warm Gas in (active) Galaxy Nuclei

1 1 Fig. 6. SMA CO(3–2) integrated map of NGC 4826 (grey scale given in the color bar in Jy km s− beam− )overlaidwiththePdBICO(1–0) 1 1 The Astrophysical Journal,745:65(14pp),2012January20 Ueda et al. integrated map (left)withcontoursat2,4,8,12,16,20,and24Jykms− beam− and the PdBI CO(2–1) integrated map (right)withcontours 1 1 at 4.1, 14.3, 24.5, 34.7, 44.9, 55.1, 65.3 and 75.5 Jy km s− beam− .Noprimarybeamcorrectionhasbeenapplied.Allmapshavethesame CO(3-2)/(1-0) resolution (PdBIA&A data 525, have been A18 tapered) (2011) and the beam (2.6 1.9##)isshownatthe bottom left.ThewhitecrosslabelledCshowsthepositionof h m s × Table 3 the dynamical center (αJ2000 = 12 56 43 .64, δJ2000 = 21◦40#59 ##. 3). The other crosses indicate the positions of the spectra plotted in Fig. 8. The Mean Integrated Intensity Ratio of Each Region 4.1. NGC 4569 NGC4569 TheNGC4826 ratio maps with the corresponding error (standard deviation) Region Mean Ratio maps are shown in Fig. 9. R21 is distributed in the range 0.2 to 0.9 with a weighted mean of 0.63. The uncertainties on R21 NGC 4038 0.6 0.2 are in the range 0.01 to 0.05. R32 is distributed in the range 0.05 ± to 0.45 with a weighted mean of 0.23. The uncertainties on R32 NGC 4039 0.5 0.1 ± are in the range 0.03 to 0.11. The errors on the weighted means Complex 1 0.4 0.1 based on the data noise only are less than 0.02. While calibration ± uncertainties introduce an overall error of the order of 0.05, they Complex 2 0.3 0.1 ± should have the same effect on all locations and thereby spa- Complex 3 0.5 0.1 tial trends should be unaffected. The mean values are in good ± agreement with the values found by Hafok & Stutzki (2003) based on single dish observations with a beam of 80"" HPBW, i.e., R = 0.64 0.08 and R = 0.23 0.04; and with the 21 ± 32 ± R32 value found by Wilson et al. (2009)withintheinner22"", et al. 2010), not considering projection effects and extinction. In i.e., 0.34 0.11 (note however that the latter is not a fully inde- pendent check± as we used the same CO(3–2) single dish data to a related study, Mengel et al. (2005)identifiedstarclustersusing complement our SMA data). the (VLT) Ks-band and found that about The R21 and R32 maps are different. While R21 is smooth, 5 5", 400 pc more or less symmetric with respect to the center, and contin- 70% of the star clusters with masses !10 M are younger than uously5", increasing100 pc toward the center with a maximum at 1 to ! ∼ "" 10 Myr and most of the young clusters with ages < 6Myrare the west of the galactic center, R32 is more clumpy, exhibiting several peaks, and is more asymmetric. The large-scale gradient located in the overlap region. Although Mengel et al. (2005) in R32 reported by Wilson et al. (2009)doesnotappearinthe inner 10 . suggested that the CO (1–0) obtained at a spatial resolution of Fig. 3. Composite color image of NGC 4569 madeFig. from 7. Composite the intensity color image of NGC 4826"" made from the intensity maps of the CO(1–0), CO(2–1) and CO(3–2) line emission. The line inten- maps of the CO(1–0), CO(2–1) and CO(3–2) linesities emission. are represented Theline by the red, greenR andis blue expected color intensities, to respectively. probe more The intensity extreme maps were conditions obtained by such summing all channels where RGB=CO1-0the line has/S2-1/N > 3andarecorrectedforprimarybeamattenuation./3-2;32 PdBI/PdBI/SMA 4## in the overlap region shows spatial correlation with the intensities are represented by the red, green and blue color intensities as those occuring in star forming or shock regions. This is ∼ respectively. The intensity maps were obtained by summing all chan- confirmed by the comparison to the Paα emission (Fig. 10). location of the star clusters, our CO (3–2) emission map with a nels where the line has S/N > 3, and are corrected for primary beam Although the detailed structure of R32 is not reflected in Paα, attenuation. A18, page 6 of 26 Boone et al. (2011) spatial resolution of 1## shows a clear offset from the star clus- most of the pixels where R32 > 0.2coincidewithPaα emission ters, assuming that the∼ star clusters identified using the V-band detected in the NICMOS image. R21,instead,reacheshighval- 10", 1 kpc peak CO(3-2)/(1-0)ues outside = (to 0.45 the north) as(N4569) well as inside the Paα emission re- image overlap with those identified using the Ks-band image. gion, although it peaks at the galactic center as the Paα.Thefact It is dominated by a nearly circular region elongated along the that R32 tracks massive star formation in NGC 4569 better than The offset is evident in the southern part of the overlap region, R 21 is consistent = 0.60 with the (N4826) conclusion of Wilson et al. (2009), galactic major axis with a radius of 2"" (40 pc) and identified where the CO (3–2) emission is strong and the CO line ratio is as a circumnuclear disk (CND) in NUGA-I∼ .Theemissionde- based on maps over larger areas, that CO(3–2) is a robust tracer tected outside the disk seems to coincide with the spiral arms of star formation. relatively high. Hence, the absence of a significant spatial corre- Figure 9. Integrated brightness temperature CO (3–2)/(1–0) ratios. The back- described in that same paper. This general agreement between The fact that R32 correlates with the individual CO line inten- lation between the locations of the molecular complexes and star ground image is the HSTpeak435 nm CO(3-2) (Whitmore et al.=2010 0.75). (N4038) the three CO lines suggests that the discrepancies in the channel sities (the three peaks in CO are also peaks in R32)alsosuggests maps are mainly due to the lower S/N of the SMA data in indi- that it is more sensitive to density. In contrast, the fact that R21 clusters may suggest that the molecular gas traced in CO (3–2) (A color version of this figure is available in the online journal.) vidual channels. This interpretation is further supported by the does not show the same structures as the individual line maps ~ 1.0 (N4039) similarities between the individual spectra shown in Fig. 8.At and that it increases steadily towards the center suggests that it emission corresponds to the regions of future star formation, or the eight positions (marked on the integratedLINER/transition map in Fig. 6)the is more sensitive nuclei to temperature. incapable This is expected for thermal- regions where star formation occurred in the past 1Myrbe- at the 1kpcscaleintheAntennaeareclosetotheaverage spectra exhibit similar profiles to within 30%. ized gas with moderate opacity in the CO(3–2) line. ∼ ∼ N4039 has less S.F. and no AGN !?The relative strengths of the linesto appear enhance to follow two R(3-2/1-0)The left panel of Fig. 11 shows the distribution of the ratios cause no that is younger than 1 Myr corresponds to properties of nearby galaxies (e.g., 0.44 is the average ratio trends. (i) The closer to the center, the stronger the CO(2–1) in the R32 vs. R21 space; in the following this will be referred the molecular complexes. This timescale is similar to the typical among seven nearbyUeda galaxies; et Mao al. et al.(2012)2010), except for two and CO(3–2) lines relative to CO(1–0); (ii) The CO(3–2) line to as the R32-vs.-R21 diagram. This diagram was obtained by is stronger relative to CO(2–1) in the central disk (positions B, mapping each pixel (0.2"" 0.2"" in size) of the ratio maps to timescale for disassociation from the molecular cloud which is localized regions where the excitation conditions are similar to CandD)andalongthemajoraxis(comparepositionsAandE the grid and convolving with× a Gaussian kernel of 0.02 FWHM, vs. F and G), where the spiral arms connect to the disk. corresponding to a typical uncertainty in the line ratio. Based under 3 Myr (Lada & Lada 2003). the central regions of starburst galaxies. on the line ratio error maps shown in Fig. 9 it should be kept Finally, we mention two issues that likely affect the analysis in mind that any feature in this diagram with a size lower than 0.05 cannot be considered as real. In addition, although the ∼ 4.2. Distribution of the Integrated and interpretation of the line ratios. One issue is the inapplica- 4. Line ratios ratio maps are considered as continuous distributions (we used Brightness Temperature Ratio small pixels with respect to the beam), it should be kept in mind bility of the Rayleigh–Jeans approximation, especially for the With the definition adopted in theintroductionanylineratio that the resolution is finite in the ratio maps. This diagram is used higher frequency CO (3–2) line. In such a case the brightness would tend to 1 with increasing excitation temperature for ther- to show in a synoptic way which regions of the ratio-ratio space We investigated the integrated brightness temperature CO malized gas in local thermal equilibrium. Line ratio maps were are populated. As expected from the differences between the R32 temperature ratio can be less than unity (0.8–0.9 for a tempera- computed restricting the spectral window of each pixel to the and R21 maps noted above, the distribution is dispersed in the (3–2)/(1–0) ratios using the SMA CO (3–2) and OVRO CO channels where the CO(1–0) flux is >4–σ and the CO(3–2) diagram. The ratios populate a region going from (0.4, 0.15) to ture of 40 K; see Harris et al. 2010), assuming a thermalized and flux is positive. When the spectral window was narrower than (0.8, 0.25) with an average thickness of 0.15 in an orthogonal (1–0) data (Wilson et al. 2000). While we assume the same 1 ∼ optically thick, single-component cloud. The other possibility 50 km s− ,i.e.,lessthan5channels,thepixelwasdiscarded. direction. beam-filling factor for both transitions, it is possible that the arises from the filling factor difference between the CO (3–2) A18, page 4 of 26 true distribution of the CO (3–2) emitting cloud is more clumpy and CO (1–0) lines. Such a multi-component cloud may be more and concentrated than the clouds emitting CO (1–0) emission. realistic than adopting a single-component cloud, and higher an- Therefore, the peak line ratios we estimate here are lower limits. gular resolution ALMA/Atacama Compact Array observations We clipped the visibilities so that both data have the same of both lines should provide us with the exact distribution of shortest uv range (minimum uv distance 10.48 kλ 9.13 m) each line tracer in the future. and then convolved the CO (3–2) data to= the angular= resolution (5##.49 3##.84) of the CO (1–0) image before estimating the line 4.2.1. The Nucleus of NGC 4038/9 ratios.× Finally, the line ratios are clipped at the 3σ level. The resultant ratio map is shown in Figure 9.Weseparatethe The mean 12CO (3–2)/(1–0) ratio of NGC 4038 is 0.6 0.2, ratio map into five regions (NGC 4038, NGC 4039, Complex 1, which is the highest among the averages of the five complexes.± Complex 2, and Complex 3) and estimate the mean ratio for The spatial distribution of the ratio presented in Figure 9 shows each region. The mean integrated intensity ratio ranges from that the ratio is 0.75 in the center, and this is two times higher 0.3 to 0.6 (Table 3). Here the error bars are derived from the than the ratios in the outer region. The regions with particularly uncertainties in the flux calibration. On average, the CO line in high line ratios correspond well with the peaks of the MIR and the Antennae is not thermalized up to J 3–2. It is slightly NIR emission, qualitatively suggesting the correlation between less than the average ratios estimated by Zhu= et al. (2003)using high line ratio and embedded star formation. This is consistent single-dish telescopes, which ranged between 0.67 and 0.91. In with observations in nearby galaxies where the ratio appears to contrast, the line ratios are close to unity in the southern part increase from the outer disk to the central regions (e.g., Israel of Complex 3 and the two nuclei. Iono et al. (2009)derived 2009;Wilsonetal.2009;Booneetal.2011). A similar trend 0.93 for the central regions of 12 dusty U/LIRGs in the local is also seen in the Milky Way, where the ratio increases from universe, which is also in agreement with the starburst galaxies 0.5intheinnerdiskto 0.9towardthegalacticcenterwhere ∼ ∼ Arp220 and M82 where the derived ratios are "1 (e.g., Greve the star formation is more active (Oka et al. 2007). Furthermore, et al. 2009). Thus, the average excitation conditions analyzed signatures of more recent star formation activity in NGC 4038

8 The Astrophysical Journal Letters,756:L10(5pp),2012September1Other Nearby AGNsEspada et al.

Molecular-gas spiral arms in dust-lane elliptical galaxy

R: 8mm PAH, Spitzer G: CO(2-1), SMA B: X-ray, Chandra

Espada et al. (2012) Figure 1. Left: optical image of Centaurus A, showing its prominent dust lane (ESO/IDA/Danish 1.5 m/R. Gendler, J.-E. Ovaldsen, & S. Guisard, ESO). Right: the molecular gas as traced by our SMA CO(2–1) observation (green), PAH and dust emission at 8 µm observed by Spitzer (red; Quillen et al. 2006), and the Chandra X-ray observations of the jet (blue, NASA/CXC/M. Karovska et al.). Note that the spiral and the 8 µmemissioninthispanelarewithintheopticaldustlanevisible in the left panel. The most prominent features of the 12CO(2–1) emission are consistent with those of the 8 µm emission, but the CO shows the spiral arms. There is a CO absorption feature toward the unresolved compact continuum component located toward the AGN (Espada et al. 2010). Note that the jet is nearly perpendicular to the circumnuclear molecular gas at the base of the spiral arms, but not to the outer component.

We present new observations using the SMA10 (Ho et al. maximum angular scale observable with the shortest baselines 2004), which allow us to shed more light into the complex few is 25!!,whichconsiderablylimitstheamountofmissingfluxto inner kiloparsecs. The resolution of the mosaic we present here emission with angular scales larger than this. is a factor of 45 higher than existing 12CO(2–1) maps covering The two tracks were reduced independently. The editing the dust lane, with a resolution of 23!! (or 380 pc; Rydbeck and calibration of the data were done with the SMA-adapted et al. 1993). Our mosaic covers an area three∼ times larger than MIR software.11 3C273 and 3C279 were used for passband previously published interferometric CO maps (Paper I)tofully calibration. An initial gain calibration was performed using cover the entire parallelogram structure. J1337-129, which is at an angular distance of 30◦ from the target. We confirm that the continuum emission toward Cen 2. SMA 12CO(2–1) OBSERVATIONS A was found to be unresolved (Paper I). The gain calibration AND DATA REDUCTION for the five pointings was then refined using the averaged line- free channels in Cen A itself (central pointing). Callisto and Cen A was observed at 1.3 mm using the SMA with seven Ganymede were used as absolute flux calibrators. Overall, we antennas on 2008 April 22 and 26. The field of view is char- estimate the absolute flux uncertainties on the order of 10%, acterized by a half-power beam width of the primary beam of from the comparison of both data sets. Finally, we combined an SMA antenna of 52!! (0.9 kpc). We observe a five point- both tracks. ing mosaic to cover 156 52!! (or 2.6 0.9 kpc). The digital An image of the 12CO(2–1) emission was produced us- × × correlator was configured with 6144 channels (2 GHz band- ing MIRIAD (Sault et al. 1995). A careful subtraction of the 1 width), resulting in a velocity resolution of about 0.5 km s− . continuum was done using line-free channels with the task The receivers were tuned to the redshifted 12CO(2–1) UVLIN.ThedatawerecleanedusingMOSMEM with uniform (νrest 230.538 GHz) emission line in the upper sideband, using weighting. The synthesized beam is 4!!.4 1!!.9(80 40 pc) = 1 VLSR 550 km s− .Notethatvelocitiesareexpressedthrough- with a major axis P.A. 25.3. The task×MOMENT was× used = ◦ out this Letter with respect to the LSR using the ra- to calculate the 12CO(2–1)= integrated flux density distribution dio convention. We used R.A. 13h25m27s.6anddecl. (2σ clipping) and the intensity-weighted velocity field distri- = = 43◦01!08!!.8(J2000)asourphasecenter(AGNposition bution (clipped at 3σ ). The rms noise level of the integrated − h m s R.A. 13 25 27.615 and decl. 43◦01!08!!.805; Ma et al. intensity map is 6 Jy beam 1 km s 1.Figure2 shows the SMA = = − − − 1998). The maximum elevation of the source at the SMA site is 12CO(2–1) integrated intensity map and velocity field along the 27◦,forcingustoobserveonlyunderverystableatmospheric inner 3 ( 3kpc). & ! conditions, with zenith opacities of typically τ225 0.10. In or- ∼ ∼ der to have a beam shape as close to circular as possible we used 3. SPIRAL ARMS IN A GIANT ELLIPTICAL GALAXY acompactconfigurationwithlongerN–Sbaselines.Minimum and maximum projected baselines were 6 m and 108 m. The From the 12CO(2–1) emission distribution in Figure 2,we confirm that the molecular gas is preferentially located along 10 The Submillimeter Array is a joint project between the Smithsonian Astrophysical Observatory and the Academia Sinica Institute of Astronomy 11 MIR is a software package to reduce SMA data based on the package and Astrophysics, and is funded by the Smithsonian Institution and the originally developed by Nick Scoville at Caltech. See Academia Sinica. http://cfa-www.harvard.edu/ cqi/mircook.html.

2 Polarimetry

SMA 10 @ Cambridge, MA, 2014 May Alignment of Molecular Cloud Magnetic Fields with Spiral arms in M33

gmc 1 gmc 2 +,-.! Li et al. (2011) )( )( )'$ !"#$&#"(%" !"#$"#$!%& !"#$1#/(%! !(

!'$ !( !( '#!$#"&%! )( CO line polarization )( '#!$#"" !( gmc 6 !01 gmc 5 )0" or with B )'" '#!!#/&%" !'$ !01

)'" )( gmc 4 !"#$*#$(%(

!'" !"#$*#'&%( !"#!!#$$%( !'" !0" !(

)( '#!$#'!%( !( )0" )'" '#!$#!'%/ '#!$#!$%/ !'" !0"

&" GMC 1 GMC 2 GMC 3 GMC 4 % GMC 5 GMC 6

$

< Correlation with arm !

#

" !!" !#" " #" !" $" %" &"" &#" &!" '(()*+,'(,-.,/'01,(2'3,0'450,523,672*4+7'8,96*:2**);, Mass Accretion Rate to M87 BH from Faraday Rotation Measure –5– –6– Kuo et al. (2014)

34 Fit RM=-2.1x105 +/- 1.8x105

33

32

31

30

29 EVPA (degrees)

28

27

26

25 1.65 1.7 1.75 1.8 1.85 1.9 2 2 λ (mm )

Fig. 2.— Total intensity of M 87 at the 232.3 GHz band inFig. grey scale. 3.— RM The orange fit at segments the center show of the the M 87 core based on polarization position angles measured at four magnetic vector position angles (MVPA) derived for the same band.differentThe MVPAs frequencies. were obtained The best-fit by rotating RM is −(2.1±1.8)×105 rad/m2. The error bars are derived from the 1σ ◦ EVPAs by 90 . The clumps from left to right in the figure areimage the nucleus uncertainty (core), j ofet the knot Q A, and B, and U maps. C, -4 respectively. The synthesized beam is shown in theRM bottom right⇒ of tdM/dthe image. < 9.210 M/yr

4. Discussion Based on Equation (9) in Marrone et al. (2006), we express M˙ as a function of RM as

− 4.1. Constraints on Mass Accretion Rate and its Accretion Flow Model −2/3 M 4/3 2 2/3 ˙ × −9 − −(3β−1)/2 × BH 7/6 2/3 2 M =2.2 10 1 (rout/rin) 9 rin RM rad/m , To constrain M˙ with RM, we essentially follow the procedure outlined in Marrone! et al. (2006). The " #6.6 × 10 M" $ #3β − 1$ primary assumption in this method is that the hot accretion flow in front of a source of synchrotron emission causes Faraday Rotation. The model also assumes that the densitwherey profileβ is a of parameter the accretion depending flow follows on a the accretion flow models (β=1/2 − 3/2 accounts for all varieties of − power-law distribution (n ∝ r β ) and that the magnetic fieldaccretion is well ordered, models). radial In and the of equipartit above equation,ion rin and rout are the inner and outer edge of the Faraday screen strength. (in units of rs) where the electrons are sub-relativistic and the magnetic field is coherent. We remark that In the case of M87, we further assume that the innermostthejet electrons (i.e. its base) within providesrin make the dominant a small or negligible contribution to RM because they become relativistic (i.e. 9 2 background polarized emission in the submillimeter band (see FigureTe >T 4)Rel. Note=6 that× 10 whileK= polarizedmec /k emissionB) and RM is, thus, suppressed (Quataert & Gruzinov 2000). could be emitted along the jet all the way from close to the BH to radii beyond rB, the dominant polarized emission at 230 GHz most likely originates from the jet base. This isIn bec orderause the to derivemajorityM˙ ofwith our SMA the flux above equation, it is important to determine rin and rout. We estimate rin density (1.6 Jy) can be explained by the emission from the jetbased base on at the the center observed of M87. flux Based density on of the M87 from VLBI observations. As we described above, thermal electrons VLBI observation of M87 at 230 GHz, Doeleman et al. (2012) show that the jet base has a flux density of within rin become relativistic and can emit significant synchrotron radiation that dominates the emission 1 Jy and a size of 5.5rs. By using the core-shift relationship (Hadafrom et the al. 2011),accretion Nakamura disk (Yuan&Asada(2013) et al. 2003). Since it is expected that a hot accretion flow would be optically determine the offset of the jet base from the BH at 230 GHz to be 4.3rs (in projected distance). Therefore, thick at cm wavelengths, the size of the region with relativistic can be constrained by using the these studies imply that the dominant background emission quite likely originates from within a few rs from the BH. Nonetheless, we currently cannot exclude the possibilityobserved of polarized flux emissions density from and the brightness diffuse part temperature. For M 87, the observed total flux density of the nucleus of the jet, and this issue will be addressed with future high-resolutis 1ion Jy observations. from the VLBI observation at 43 GHz (Abramowski et al. 2012). Since the brightness temperature 9 of the hot electrons within rin must be at least 6×10 K, the diameter (i.e. 2 rin) of the emitting region of relativistic plasma can be constrained to be ≤ 0.32 mas (= 42rs). Since the observed flux density sets an upper limit on the amount of emission from relativistic electrons, it leadstoanupperboundonrin of 21rs. On the other hand, if the innermost jet resides right in front of the the emitting region of the relativistic plasma and the emission from the hot electrons is totally absorbed by the optically thick jet, we can not constrain the size of the region of the relativistic plasma directly fromtheobservedflux.However,wecan use the size of the innermost jet to constrain rin. The observed size of the VLBI core at 43 GHz is 17 ± 4rs (Asada & Nakamura 2012). If the emission from the hot electrons is totally absorbed, rin must be smaller than 17 ± 4rs.Therefore,the21rs upper bound for rin is reasonable even for this case. We fix rout to be at the Bondi radius rB.Here,rB is just a convenient value to choose while rout is not well determined. This is SMA on Nearby Galaxies • Spectroscopy - new masers, vib-excited lines • U/LIRG Legacy Project - gas concentration → luminosity @ constant dense gas SFE • U/LIRGs at highest resolutions - new AGN/Starburst diagnosis with sub-mm subarcsec obs. • Feedback/Outflows - new cases, forms, detection methods • Warm/Dense Gas in AGN - AGN effects on e.g.,high-J CO • Polarimetry - Magnetic filed in MCs, AGN accretion • ... Many discoveries on warm/active sources Ongoing follow-ups with SMA, ALMA, etc.

SMA 10 @ Cambridge, MA, 2014 May SF in Arp 220 and NGC 4418

EVN+MERLIN 5GHz 0.02" res. -00 52 39.30 0.4 0.3 H 0.2 39.35 D 0.1 A

0.0 39.40 B -0.1 S F C Dec. offset [arcsec] G -0.2 N 0.25"x0.21" -0.3 Declination (J2000) 100 pc E 39.45 E -0.4 0.8 0.6 0.4 0.2 0.0 -0.2 -0.4 -0.6 -0.8 -1.0 R.A. offset [arcsec] SMA submm 39.50 core size

red + : VLBI SNRs & RSNe 12 26 54.624 54.620 54.616 54.612 54.608 Right ascension (J2000) (Lonsdale et a. 2006; Parra et al. 2007) clumps (SSCs) with contours : SMA 860μm continuum D ~ 5 pc, Tb ~105.2±0.2 K (Sakamoto et al. 2008) ( + AGN ??) (Varenius et al. 2014) SF evident. Is AGN hidden !?