The CO Emission in the Taffy Galaxies (UGC 12914/15) at 60 pc Resolution. I. The Battle for Star Formation in the Turbulent Taffy Bridge

The CO Emission in the Taffy Galaxies (UGC 12914/15)at 60 pc Resolution. I. The Battle for Star Formation in the Turbulent Taffy Bridge P. N. Appleton 1 , B. Emonts 2 , U. Lisenfeld 3 , E. Falgarone 4 , P. Guillard 5 , F. Boulanger 6 , J. Braine 7 , P. Ogle 8 , C. Struck 9 , B. Vollmer 10 , and T. Yeager 11 1 Caltech/IPAC, MC 314-6, 1200 E. California Boulevard, Pasadena, CA 91125, USA; [email protected] 2 National Radio Astronomy Observatory, 520 Edgemont Road, Charlottesville, VA 22903, USA 3 Departamento de Fisica Teorica y del Cosmos, Universidad de Granada, Spain and Instituto Carlos I de Fisica Teorica y Computacional, Universidad de Granada, Spain 4 LERMA/LRA, Observatoire de Paris, PSL Research University, CNRS, Sorbonne Universites, UPMC Universite Paris 06, Ecole normale superieure, F-75005 Paris, France 5 Sorbonne Universités, UPMC Paris 6 et CNRS, UMR 7095, Institut d’Astrophysique de Paris, 98 bis Bd Arago, 75014 Paris, France 6 UPMC Universite Paris 06, Ecole normale superieure, F-75005 Paris, France 7 Laboratoire d’Astrophysique de Bordeaux, Univ. Bordeaux, CNRS, B18N, allée Geoffroy Saint-Hilaire, F-33615 Pessac, France 8 Space Telescope Science Institute, 3700, San Martin Drive, Baltimore, MD21218, USA 9 Dept. of Physics and Astronomy, Iowa State University, Ames, IA, 50011, USA 10 Université de Strasbourg, F-67000 Strasbourg, France 11 Lawrence Livermore Laboratories, 7000 East Avenue, Livermore, CA 94550, USA Received 2021 November 4; revised 2022 March 29; accepted 2022 March 29; published 2022 June 1 Abstract We present Atacama Large Millimeter/submillimeter Array observations at a spatial resolution of 0 2(60 pc)of CO emission from the Taffy galaxies (UGC 12914/5). The observations are compared with narrowband Paα, midIR, radio continuum and X-ray imaging, plus optical spectroscopy. The galaxies have undergone a recent head-on collision, creating a massive gaseous bridge that is known to be highly turbulent. The bridge contains a complex web of narrow molecular filaments and clumps. The majority of the filaments are devoid of star formation, and fall significantly below the Kennicutt–Schmidt relationship for normal galaxies, especially for the numerous regions undetected in Paαemission. Within the loosely connected filaments and clumps of gas we find regions of high velocity dispersion that appear gravitationally unbound for a wide range of likely values of X CO . Like the “Firecracker”region in the Antennae system, they would require extremely high external dynamical or thermal pressure to stop them dissipating rapidly on short crossing timescales of 2–5 Myr. We suggest that the clouds may be transient structures within a highly turbulent multiphase medium that is strongly suppressing star formation. Despite the overall turbulence in the system, stars seem to have formed in compact hotspots within a kiloparsecsized extragalactic H II region, where the molecular gas has a lower velocity dispersion than elsewhere, and shows evidence for a collision with an ionized gas cloud. Like the shocked gas in the Stephan’s Quintet group, the conditions in the Taffy bridge shows how difficult it is to form stars within a turbulent, multiphase, gas. Unified Astronomy Thesaurus concepts: Interacting galaxies (802) Supporting material: machine-readable table 1. Introduction Major mergers between massive gas-rich galaxies are transformative events in galaxy evolution. In many mergers, models suggest that tidal torques within host galaxy disks drives gas inward, often forming an intense dust-enshrouded nuclear starburst (e.g., Mihos & Hernquist 1996). Such a mechanism has been long accepted as the main explanation for the existence of rare, but powerful sources of far-IR (FIR) emission (LIRGs and ULIRGs) 12 in the local universe (e.g., Soifer et al. 1984,1984; Sanders et al. 1986; Armus et al. 1987; Sanders & Mirabel 1996; Armus et al. 2009; Saito et al. 2015; Armus et al. 2020). While much of the early modeling of colliding and massive merging galaxies primarily was aimed at more general collision geometries, head-on collisions, thought to be responsible for collisional ring galaxies, have always been a special case (Lynds &Toomre1976;Toomre1978; Appleton & Struck-Marcell 1987; Struck-Marcell 1990;Gerberetal.1996). Although several of these systems involved dissipative gas-rich collisions (e.g., Appleton et al. 1996; Charmandaris & Appleton 1996;Higdon 1996;Braineetal.2003,2004), there has been a resurgence of interest in the treatment of the dissipative effects of the gas (Renaud et al. 2018), especially the formation of a “splash”bridge (Struck 1997;Yeager&Struck2019,2020a,2020b). Splash gas bridges are produced when two gas-rich disk systems collide nearly head-on. In such cases, the stellar components pass through each other, but leave behind a massive gas bridge (Vollmer et al. 2012). These kinds of collisions are challenging to models because a large fractions of the gas is strongly compressed and heated during the collision, and much of the multiphase medium remains at various stages of cooling tens of millions of years after the collision (e.g., Yeager & Struck 2020a,2020b). These later models suggest that the gas between the galaxies continues to The Astrophysical Journal, 931:121 (23pp), 2022 June 1 https://doi.org/10.3847/1538-4357/ac63b2 © 2022. The Author(s). Published by the American Astronomical Society. Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence. Any further distribution of this work must maintain attribution to the author(s)and the title of the work, journal citation and DOI. 12 (U)LIRG =(ultra)luminous infrared galaxies in the local universe are defined as having high far-IR luminosities L IR >(10 12 L e )10 11 L e . 1 collide with high Mach-number, generating strong turbulent conditions in the bridge. Gas in such bridge systems can be used to study the effects of turbulence and shocks on the formation of stars in a relatively “clean”environment, far from the complicating effects of nuclear starbursts or active galactic nuclei (AGNs) foundinmanyothermajormergersystems. We present new high-resolution CO observations of one of the best studied splash bridge systems, known as the Taffy galaxies (UGC 12914/5; Condon et al. 1993; stellar masses of 7.4 and 4.2 ×10 10 M e , respectively; Appleton et al. 2015). The two gas-rich disks are believed to have collided almost face-on at high velocity (600–800 km s −1 )25–30 Myr ago (see Figure 1(a)). The ionized gas in the galaxy disks appear to be counterrotating (Joshi et al. 2019), suggesting the disks had oppositely oriented spins when the disks first collided. This geometry would create an even stronger affect on the gas collisions at the time of impact. We are now likely viewing the system almost edge-on after the stellar components have passed through each other (Vollmer et al. 2012,2021), leaving behind a massive neutral and molecular gas bridge in the center of mass frame of the galaxies. In addition to the disturbed CO distribution studied with varying degrees of spatial resolution (Gao et al. 2003;Zhuetal. 2007;Braineetal.2003; Vollmer et al. 2021),thereis independent evidence that the gas in the bridge is highly disturbed. Spitzer InfraRed Spectrograph (IRS)observations have shown the existence of large amounts of warm (T∼100–300 K) emitting molecular hydrogen (Peterson et al. 2012)with properties consistent with shock or turbulent heating (Guillard et al. 2009). The spectra showed powerful dominant emission lines of pure-rotational H 2 , with large ratios of warm H 2 /PAH and H 2 /FIR, similar to those seen in the Stephan’s Quintet intergalactic shock (Appleton et al. 2006;Cluveretal.2010). Herschel PACS observations showed that the bridge also emits strong [CII]158 μmand[OI]63 μm emission, as well as emission from [CI],CO(4-3),andCO(5-4)based SPIRE FourierTransform Spectrometer observations (Peterson et al. 2018).The strength and unusual line ratios of the fine structure lines point toward heating by shocks and turbulence, as suggested by recent models (Yeager & Struck 2020b). Direct evidence for fast atomic shocks (V∼200 km s −1 )was found in the bridge (Joshi et al. 2019), as well as potentially faster shocks from Chandra observations of soft X-ray emission (Appleton et al. 2015). Radio continuum emission from the bridge (Condon et al. 1993) can also be explained as a result of Fermi accelerations of cosmic rays in shocks generated within the turbulent gas (Lisenfeld & Völk 2010). Recently, Vollmer et al. (2021)reported high spatial resolution (27)observations of CO (1-0)in the Taffy system with the IRAM Plateau de Bure Interferometer (PdBI),aswellas detailed models of the structure and kinematics of the large-scale gas distribution. The results provide very strong support for the collisional picture, and evidence for star formation suppression in the bridge. The results suggest that much of the gas in the bridge is not virialized on the 800 pc to few kiloparsec scales, and they suggest turbulent adiabatic compression is responsible for the high velocity dispersion in the observed gas clumps. One exception is a luminous extragalactic H II region (hereafter X-H II region), which may have formed in the bridge, close to the northern-most Taffy galaxy UGC 12915. This paper presents Atacama Large Millimeter/submillimeter Array (ALMA) 13 observations of the CO (2-1)emission Figure 1. The Taffy galaxy system, (a)shown in a false-color Sloan Digital Sky Survey (SDSS)bri-band (blue =b, green =r, red =i)image designed to show the faintest emission, including a faint extragalactic H II region just southwest of UGC 12915; (b)red circles show the primary beams of the 37 Atacama Large Millimeter/submillimeter Array (ALMA)12 m pointings (FWHM =25 6)for CO (2-1)observation described here. 1″corresponds to 300 pc at D=62 Mpc. 13 ALMA, an international astronomy facility, is a partnership of ESO, the U.S. National Science Foundation (NSF)and the National Institutes of Natural Sciences (NINS)of Japan in cooperation with the Republic of Chile. 2 The Astrophysical Journal, 931:121 (23pp), 2022 June 1 Appleton et al. from the Taffy system on angular scales (0.24 ×0.18 arcsec 2 ), an order of magnitude higher than previous work. Because of the large quantity of data obtained in the ALMA mosaics, we split our discussion of the ALMA CO data into two papers. The current one concentrates on the CO (2-1) emission from the gas bridge from Cycle 4. A second paper will discuss more fully the condition of the molecular gas in the two Taffy galaxies themselves, and will present CO (3-2)data obtained both in Cycle 4, and in Cycle 7. The main goals of this paper are, (1)to explore the distribution of molecular gas in the Taffy bridge at 60–100 pc resolution and its relationship to Paαemission observed with HST at similar resolution, (2)probe the kinematics of the CO gas emission and how that relates to shocks and star formation previously observed from groundand space-based data, and (3)study in greater detail the one major area of star formation in the bridge (the extragalactic H II region). This region may provide further insights into the formation of stars in turbulent environments through comparison with new optical spectroscopy, new radio continuum, and archival Spitzer and X-ray observations. The paper is organized as follows. The observations and data calibrations are described in Section 2. Results, including the large-scale molecular distribution and kinematics and gas surface densities estimates are presented in Section 3. Section 4 describes the ionized gas emission in the bridge and in the X-H II regions. Section 5is concerned with the relationship between molecular gas and star formation in the bridge, including testing the Kennicutt–Schmidt relationship for the clouds, the virial properties of the clouds, and quantifying the star formation rate (SFR)in the X-H II region. The Taffy bridge is compared with other similar intergalactic environments in Section 6. The possible origin of large-scale star formation suppression in the bridge is described in Section 7. The conclusions are given in Section 8. The Appendix includes tabulated data and figures relating to the extracted regions discussed in the main body of the paper. We assume a distance to the Taffy galaxies of 62 Mpc based on a mean heliocentric velocity for the system of 4350 km s −1 , and a Hubble constant of 70 km s −1 Mpc −1 . At this distance, 1″ corresponds to 300 pc. 2. Observations 2.1. CO Observations and Reduction Table 1is a complete list of the observations made with the ALMA 12 m arrays in programs 2016.1.01037.S (40 antennas) and 2019.1.01050.S (41 antennas). In 2016.1.01037.S, 37 fullsampled primary beam pointings with the 12 m array were made of the Taffy pair and bridge in 12 CO (2-1)(ν rest =230.54 GHz; see Figure 1(b)). Two sets of seven pointings were also made in the 12 CO(3-2)(ν rest =345.795 GHz)centered on the brightest part of the bridge, and a bright region in UGC 12914. A further set of CO (3-2)observations were made of the bridge and UGC 12915 in program 2019.1.01050.S (41 antennas). Although we list all of the observations in the table, we postpone discussion of the CO (3-2)observations until a second paper. The CO (2-1)observations centered on a heliocentric (optically defined)velocity of 4450 km s −1 were observed with a total bandwidth of 1.875 GHz (2472 km s −1 )and a channel separation of 1.953 MHz (2.6 km s −1 )in ALMA Band 6. A second continuum baseband was centered on ν=228 GHz (λ1.3 mm, and bandwidth 1.875 GHz). Calibration of the CO (2-1)data was performed using ALMA flux and phase-reference calibration sources during the course of the observations. These data were processed with a standard ALMA calibration pipeline included in CASA v.5.5.1-5 resulting in fully flux, phase and bandpass-calibrated visibility data. The quality of the calibration was carefully reviewed before performing exploratory Fourier transforms of the ALMA visibility data to produce initial “dirty channel maps”smoothed to a resolution of 10 km s −1 . The maps were made in each channel over a scale of 8000 ×8000 pixel 2 , where the pixel scale was 0 018. Extended CO emission was suspected from the galaxies and weaker emission from the bridge, with emission being present not only at the smallest scales sampled by the ALMA observations (022 ×0 18)but also on slightly larger scales. This became clear when we initially tried a conventional CLEAN method (e.g., Högbom 1974)to deconvolve the “dirty maps”using the interferometric point-source response, or “dirty beam,”for each channel where emission was detected. Because of the extended emission, this procedure always led to poor negative large-scale residual flux (bowls)in some of the Table 1 Detail of ALMA Observations Cycle Band Array Date Line Frequency Mosaic Total Time Final Synthesized Refs. a Sky (GHz)(number)Time (min)On-source (min)Beam Size (″) 4 6 12 m C40-6 2016-10-04 CO(2-1)227.1 37 b 63.3 39.6 0.24 ×0.14 1 4 6 12 m C40-6 2016-10-05 CO(2-1)227.1 37 b 64.2 39.6 0.24 ×0.14 1 4 6 12 m C40-6 2016-10-06 CO(2-1)227.1 37 b 72.97 39.6 0.24 ×0.14 1 4 6 12 m C40-3 2016-12-03 CO(2-1)227.1 37 b 59.6 39.6 0.24 ×0.14 1 4 7 12 m C40-5 2016-10-27 CO(3-2)340.8 2-7 c 96.5 53.4 0.23 ×0.18 2 7 7 12 m C43-5 2021-06-30 CO (3-2)340.8 14 d 197.4 108.6 0.21 ×0.18 2 Notes. a 1=this paper, 2 =Paper IIP. N. Appleton et al. (2022, in preparation). b 37 primary beam positions were observed, fully sampling the Taffy system; see Figure 1(b). c Seven well-sampled primary beam positions were observed in two regions, one centered on the Taffy bridge, and a second in the southeast disk of UGC 12914 (to be described more fully in a second paper). d 41 antennas, 14 pointings covering bridge and inner regions of UGC 12915. 3 The Astrophysical Journal, 931:121 (23pp), 2022 June 1 Appleton et al. residual maps. Instead, we used the multiscale CLEAN algorithm (hereafter MSCLEAN; Cornwell 2008)as implemented in the CASA task “tclean.”Unlike conventional CLEAN methods, which assume that the intrinsic brightness distribution of sources is made up of points sources (corresponding to a set of zero-scale delta-functions), the multiscale clean method allows for both point-sources and larger scales to be present. Conventional CLEAN deconvolution methods iteratively build up the source distribution out of deltafunctions, by subtracting the dirty beam from the observed dirty maps (the subtraction is usually done in the uv-plane). MSCLEAN chooses from a set of smoothed dirty beams (the point-source response convolved with the MSCLEAN scale), and progressively builds a model of the large-scale flux first, followed by flux on smaller and smaller scales, until it approximates that of a point source (zero-scale). This deconvolution of emission on different spatial scales decreases low-level artifacts caused by the point-spread function (PSF), allowing us to better recover the extended emission in the image. A full description of the method can be found in Cornwell (2008). Illustrative examples of its application to cases of nearby galaxies with extended HI emission are provided by Rich et al. (2008). In our case, we tested, by trial and error, various scales to optimize the removal of the negative features seen previously in the residual maps. These final scales used corresponded to scales of zero (delta function), 6, 12, and 24 pixels. The method was then applied to all of the channel maps (10 km s −1 separation, covering a heliocentric optical velocity range from −419 to +388 km s −1 centered at 4350 km s −1 )containing emission, leading to a large data cube of the CO(2-1)emission covering a large part of the Taffy field (see Figure 1(b)). As presented in Table 1, the maps resulted in a synthesized beam with an angular resolution of 0.24 ×0.14 arcsec 2 , which corresponds to a projected physical scale of ∼60 pc for the Taffy system. The rms noise in each channel map was ∼0.7 mJy beam −1 . 2.2. HST PaαObservations Hydrogen recombination lines, like Paα, trace ionized gas associated with star formation, ionized shocks and other sources of diffuse ionized gas, and are commonly used to estimate star formation rates (e.g., Kennicutt 1998a; Calzetti et al. 2007).In Section 4we will discuss how the star formation rates estimated from the Paαemission must also take into account strong contamination from shocked gas (Joshi et al. 2019).These authors did not find evidence for other sources of diffuse emission (e.g., DIG; see Haffner et al. 2009)because of the low star formation rates in the system. We use archival Hubble Space Telescope (HST)Near Infrared Camera and Multi-Object Spectrometer (NICMOS)observations taken in the F187N and F190N filters, which were centered on UGC 12915 from the archive. These observations involved a small mosaic covering an area of 45.7 ×45.3 arcsec 2 centered on the galaxy. We subtracted the continuum images to obtain a Paαimage. The NICMOS-NIC3 image extends over part of the northern bridge, allowing us to compare the emission-line image with the CO map. Another NICMOS image of UGC 12914 is also available, but it does not cover any significant part of the bridge and is not presented here. The NICMOS observations are obtained on a 0.2 arcsec pixel scale, which slightly under-samples the PSF at this wavelength (FWHM =025 at 2 μm). This resolution is comparable to that of the CO (2-1)ALMA observations. Since the absolute astrometry of the NICMOS images is known to have significant uncertainty, we aligned the World Coordinate System (WCS)coordinates of the well-defined nuclear peak of the galaxy in the F190N filter to that of a 6 GHz radio continuum image obtained with the Karl G. Jansky Very Large Array (VLA; P. N. Appleton 2022, in preparation)by reregistering the NICMOS image (a total shift of 0 9). This offset was confirmed by finding good agreement between several compact 6 GHz radio sources, and corresponding bright compact Paαknots in the western disk of UGC 12915. Similarly, we were also able to confirm the new coordinates by comparing the position of two radio hotspots embedded in the X-H II region with the corresponding knots of star formation at the same position in the Paαimage. From these tests, we believe that the astrometry in the Paαimage is accurate to ±01(1/2 NICMOS pixel). Paαflux densities were calculated using the conversion from counts s −1 pixel −1 to erg s −1 cm −2 arcsec −2 evaluated using the relation F line =1.054 ×PHOTFLAM ×FWHM (in erg s −1 cm −2 )where PHOTFLAM is the photometric calibration parameter obtained from the observation FITS metadata, and FWHM is that of the filter. 14 The range of detected emission has a surface brightness of 1.5 ×10 −16 <Σ(Paα)<4.3 ×10 −15 erg s −1 cm −2 arcsec −2 , with a median signal-to-noise ratio (S/N)of 35. Upper limits were calculated by 2.5×the rms ADU/pixel over sample areas in the vicinity of the filaments. This upper limit, when converted to surface brightness units is Σ(Paα)<4.3 ×10 −17 erg s −1 cm −2 arcsec −2 . 2.3. Radio Continuum Observations Radio continuum observations provide another means of estimating star formation through thermal and nonthermal processes associated with active star formation regions (Condon 1992; Murphy et al. 2011). Deep observations were made as part of a radio polarization study of the Taffy (project 19A-378)in the A-array of the VLA at 1.4 GHz (Lband)and 6 GHz (Cband)during a 12.6 hr period in 2019 August 10. Although the main aim of the project was to measure radio polarization, total intensity maps (Stokes I)were made in the two bands after processing with the task “tclean”(using CASA v5.6.1)resulting in maps with restored synthesized beams of 0.36 ×0.30 arcsec 2 (Cband)and 1.2 ×1.1 arcsec 2 (Lband). The rms noise in each map was 2 μJy beam −1 and 7 μJy beam −1 in the Cand Lbands, respectively. A more complete discussion of the radio observations will be provided in a future paper (P. N. Appleton 2022, in preparation). 2.4. Chandra X-Ray Observations We make use in this paper of a Chandra X-ray (0.5–8 kev) point-source image discussed in greater detail in Appleton et al. (2015), and made available to us by those authors. The image was obtained in 2013 from a 39.5 ks exposure onto the back-illuminated S3 Advanced CCD Imaging Spectrometer (Weisskopf et al. 2000), and was smoothed with a Gaussian of FWHM 1 5 to emphasize the compact structure. In Section 4.2, we discuss the relationship between the VLA radio, HST Paα, and Chandra X-ray observations of the extragalactic H II region. Because the Chandra X-ray observations had poor absolute astrometry, we used the VLA radio 14 Based on the metadata associated with the observations, PHOTFLAM = 3.33 ×10 −18 erg s −1 cm −2 Å −1 DN −1 , and the FWHM =147.6 Å 4 The Astrophysical Journal, 931:121 (23pp), 2022 June 1 Appleton et al. map of the nucleus of UGC 12914 (a point source as observed by both instruments)to carefully register the Chandra image 0.5–8 kev image, made available to us by those authors, to the radio position. The re-registration of the Chandra image to the VLA WCS frame resulted in a shift of 0 4. This decisively shows that the brightest ultraluminous X-ray (ULX)source in the Taffy system falls within the envelope of the high surface brightness Paαand radio continuum emission from the extragalactic H II regions. A full discussion of the observations can be found in Appleton et al. (2015). 2.5. Spitzer Archival Observations IR emission is emitted from the dust grains heated by photons from young massive stars in star-forming (SF) regions. In Section 4.2 we provide IR properties of the extragalactic H II region derived from archival Spitzer observations in the IRAC (Fazio et al. 2004)3.6, 4.5, 5.8, and 8 μm bands, and in the MIPS (Rieke et al. 2004)24 μm band. Images were dearchived from the Infrared Science Archive held at IPAC, Caltech. For the IRAC and MIPS images, the photometry reported in Section 4.2 for the X-H II region was obtained by measuring flux densities in a fixed aperture of radius 4 8 centered on the clearly resolved emission from the source. The same aperture was used for the 24 μm data. However, because of the location of the X-H II region close to the disk of UGC 12915, care was taken to obtain local background estimates parallel to the extended disk of the galaxy. This effect was less than a few percent for the IRAC images, but represented a larger source of uncertainty for the MIPS 24 μm emission. These uncertainties are reflected in the photometry reported in Table 2. 2.6. Palomar 5 m Spectroscopic Observations Observations of the Taffy System were made in moderate seeing conditions (∼1″), and a 1″wide slit with the Double Beam Spectrograph (DBSP; Rahmer et al. 2012)of the Palomar 5 m telescope on 2021 January 9. The 600/1000 grating was used on the red arm of the spectrograph. At Hα, the spectral resolution was 85 km s −1 , and the scale along the slit was 0.29 arcsec/pixel. Flux calibration was performed using short observations of the white dwarf star G191B2B. The total on-source integration time was 3000 s. These data were reduced using a Python-based pipeline PypeIt (Prochaska et al. 2020), which performed bias, dark subtraction, flat field correction (using dome flats), and flux and wavelength calibration using internal lamps. 3. Results 3.1. Large-scale Molecular Gas Distribution Early CO (1-0)observations by Gao et al. (2003)of the Taffy system not only detected gas in the galaxies, but also measured large quantities of molecular gas in the bridge (1.4 ×10 10 M e , assuming a standard Galactic N(H 2 )/I CO conversion factor 15 ). However, this mass is likely to be greatly overestimated (Braine et al. 2003; Zhu et al. 2007)because there is evidence that a much smaller N(H 2 )/I CO conversion factor is appropriate in the bridge. Higher-resolution observations of the system with a beam size 2.7 x 2.7 arcsecs 2 were obtained by Vollmer et al. (2021)with the PdBI, and the integrated map is presented in Figure 2(a)superimposed over a Sloan Digital Sky Survey (SDSS)image. The observations show the bridge is composed of partially resolved clumps scattered between the galaxies. In the current paper, our ALMA CO (2-1)observations, which have a spatial resolution 50 times higher than the BIMA observations, and 12 times that of the PdBI, are shown in Figure 2(b). This full-resolution moment-0 map was constructed in the following way: (i)each channel map (of width 10 km s −1 )was spatially smoothed to an effective resolution of 0.4 ×0.4 arcsec 2 ,(ii)a valid mask was constructed of all signals within the smoothed map that was >3.5σabove the noise per channel, (iii)an integrated (moment-0)map was then made by applying the masks to the full-resolution channels, and summing the emission spatially in those channel maps (velocity space)where the mask indicated valid points above the masking threshold. Regions outside the valid regions were not summed. The same mask was also applied to the smoothed version of the cube, resulting in a smoothed moment-0 map, which is presented in Figure 3(a). This figure shows some of the fainter features better than the full-resolution image with 8000 ×8000 pixel 2 , which is not well represented in a small figure. However, the full-resolution map is used for the majority of the analysis. The ALMA observations show that many of the features, seen in the lower-resolution PdBI map, appear as narrowly defined filaments and small clumps. In addition to the bridge regions, which are the main focus of this paper, most of the detected ALMA emission from UGC 12914 is seen in narrow structures, including peculiar narrow ripples of emission along its northwestern disk, with fainter gas filaments breaking off from the disk into narrow strands, which point almost perpendicularly to the main arc of the emission in that arm. Narrow dense structure is also seen on the southeastern part of the same disk. The region around the nucleus of UGC 12914 contains high surface brightness emission distributed in a series of loops and spiral filaments (see inset in Figure 2(b)). On the other hand, UGC 12915, which is more edge-on than its companion, is dominated by a bright southeastern curved structure, or possible tightly wound spiral feature, and a high surface brightness inner edge-on nuclear disk (see upper inset in Figure 2(b)). The main galaxy disk extends to the northeast following the inner optical dust lane where it breaks up into numerous clumps and extended filaments to the far northeast. Narrow filaments of gas are also seen extending away from many parts of UGC 12915ʼs disk in strands to the north. Because of the complexity of the system, we will mainly concentrate on the Taffy bridge in the current paper, and will return to a more complete description of the CO emission from the galaxy disks in Paper II. The structure of the bridge emission is striking. The region of the bridge closest to the northern Taffy galaxy is composed of a collection of filaments and bright clumps of emission, some of which give the appearance of a cone-shaped structure whose apex lies ∼10″southwest (3kpc)from the center of UGC 12915. However, other fainter filaments cross the structure (Figure 3), and we will show that the velocity structure of the bridge gas is quite complex, and the coneshaped pattern does not form a single coherent kinematic structure. The faint X-H II region, seen in the optical image in 15 Assuming a Galactic conversion factor N(H 2 )/I CO =2×10 20 cm −2 (Kkms −1 ) −1 . 5 The Astrophysical Journal, 931:121 (23pp), 2022 June 1 Appleton et al. Figure 2(a), lies buried in the tangled northwestern part of the main emission from the bridge. There are many other scattered CO complexes apparent in the observations, including gas to the far northwest of UGC 12915. Some of this gas may be part of a separate “northwestern”bridge of emission identified in optical Integral Field Unit (IFU)observations of ionized gas in the Taffy bridge (Joshi et al. 2019). Previous observations of both the neutral hydrogen (Condon et al. 1993)and Herschel [CII]158 μm emission (Peterson et al. 2018)show neutral gas in this area. An important question to ask is what fraction of the CO emission in the Taffy bridge is detected on the small (60–100 pc) scale compared with single-dish observations (Braine et al. 2003; Zhu et al. 2007)or large-beam interferometric observations like BIMA (Gao et al. 2003). According to Zhu et al. (2007) and Gao et al. (2003), the BIMA observations of the Taffy, made with a 9.8 ×9.7 arcsec 2 beam, capture most of the flux seen in the single-dish observations. Based on a data cube provided by Y. Gao (2018, private communication),we centered a 20 5 diameter circular aperture on the main concentration in the northern bridge, and estimated the integrated flux over the measured profile to be 92.1 Jy km s −1 in CO (1-0), which agrees with a statement made in Gao et al. (2003)of emission associated with the region they called the “HII region,”which actually includes most of the structures we have been discussing. To compare this with the ALMA observations requires converting this flux to an equivalent CO (2-1)flux over the same area. Assuming r 21 =0.79 (Zhu et al. 2007; where r 21 is the ratio of I CO(2−1) /I CO(1−0) ), the equivalent CO (2-1)flux should be 291 Jy km s −1 for the Table 2 Observed and Derived Properties of the Taffy Extragalactic H II (X-H II)Region Obs/Method R ap Area Flux Density Line Flux LOG(L)SFR SFR cor LOG(Σ SFR ) (arcsec)(kpc 2 )(mJy)(erg s −1 cm −2 ×10 −15 (erg s −1 )(M e yr −1 )(M e yr −1 )(M e yr −1 kpc −2 ) [1][2][3][4][5][6][7][8][9] IRAC3.6 μm 4.8 6.4 0.38 ±0.03 L41.18 ±0.04 b LL L IRAC4.5 μm 4.8 6.4 0.32 ±0.03 L40.98 ±0.04 b LL L IRAC 5.8 μm 4.8 6.4 1.46 ±0.15 L41.54 ±0.04 b LL L IRAC 8 μm 4.8 6.4 4.33 ±0.43 L41.86 ±0.04 b LL L MIPS 24 μm 4.8 6.4 6.1 ±1.9 L41.55 ±0.13 0.08 ±0.04 c L−0.94 c NIC 3 Paα1.56 0.69 L9.1 ±1.4 39.6 ±0.06 0.20 ±0.03 0.24 ±0.04 i −0.5 (ALL) Paα sf 1.56 0.69 L5.0 ±0.29 e 39.36 ±0.06 e 0.11 ±0.03 d,d 0.13 ±0.03 i,d,e −0.72 (ALL) e Paα0.56 0.09 L2.1 ±0.2 38.97 ±0.06 0.04 ±0.01 0.05 i −0.23 (Hotspot N) Paα0.56 0.09 L3.0 ±0.45 39.14 ±0.06 0.07 ±0.02 0.08 i −0.1 (Hotspot S) Paα+24 μm 1.56 f 0.69 f LL L0.15 ±0.05 d,f L−0.66 d,f GCMS Hα4.8 6.5 L16.6 ±3.3 39.88 ±0.08 LL — (ALL) Hα sf e 4.8 6.5 L8.3 ±1.6 e 39.58 ±0.08 e 0.14 ±0.03 0.50 ±0.1 a −1.11 a Hα sf +24 μm 4.8 6.5 LL L0.09 ±0.02 f L−1.86 Radio (6GHz) 1.56 0.69 0.33 ±0.03 L[1.51 ±0.15 0.25 ±0.04 g L−0.41 (ALL)×10 20 WHz −1 ] g . Radio (6GHz) 0.56 0.08 0.052 ±0.005 L[2.38 ±0.24 0.040 ±0.002 g L−0.32 (Hotspot N)×10 19 WHz −1 ] g . Radio (6GHz) 0.56 0.09 0.086 ±0.009 L[3.94 ±0.40 0.07 ±0.01 g L−0.11 (Hotspot S)×10 19 WHz −1 ] g . Radio (1.4GHz) 1.56 0.69 1.2 ±0.1 L[5.17 ±0.50 0.29 ±0.03 h L−0.38 (ALL)×10 20 WHz −1 ] h . Notes. a SFR cor : extinction corrected for A v =1.5 mag based on the Balmer decrement of Joshi et al. (2019). b Luminosity νF(ν), and assuming a distance to Taffy of 62 Mpc. We used ν=8.3, 6.7, 5.2, 3.8, and 1.25 ×10 13 Hz for the IRAC bands 1, 2, 3, and 4 and MIPS 24 μm, respectively. c Assuming a simple monochromatic relationship of Relaño et al. (2007)and Calzetti et al. (2010), SFR 24μm [M e yr −1 ]=5.66 ×10 −36 ×L ν (24 μm) 0.82 , where L ν (24 μm)is in erg per second. For the star formation surface density, we assume the 24 μmflux is emitted from the same area as the Pαflux, i.e., from an area of 0.69 kpc 2 . d Assuming CASE B recombination, f(Hα)=2.86/0.332 ×f(Paα), and the SFR—Hαrelations of Calzetti et al. (2007,2010): SFR([M e yr −1 )=5.3 ×10 −42 L(Hα corr ), where L(Hα corr )is extinction corrected in erg per second. e Assuming both Paαand Hαfluxes have ∼45% contribution from shocks, so Paα sf (star formation)=0.55 ×Paα tot (Joshi et al. 2019). f Assuming SFR (measured Hα+24 μm)=5.3 ×10 −42 [L(Hα sf )+0.031L(24 μm)], and that the 24 μmflux comes from the same area as the Hαemission (Calzetti et al. 2007). g From VLA 6 GHz flux densities and luminosities from P. N. Appleton et al. (2022, in preparation)assuming the sum of thermal (T=10 4 K), nonthermal 6 GHz contributions from Murphy et al. (2011), and calculated spectral index of γ=−0.76 (this work)within a common area of 2.3 arcsec 2 area. h From VLA 1.48 GHz flux density and luminosity of P. N. Appleton et al. (2022, in preparation), and assuming the SFR prescription of Condon (1992)and nonthermal fraction of flux =0.8. The source is very extended at 20 cm compared with the restoring beam =1.27 ×. 1.1 arcsec 2 . i Assuming 0.2 mag of extinction at Paα. 6 The Astrophysical Journal, 931:121 (23pp), 2022 June 1 Appleton et al. same aperture. From our ALMA observations, we derived, by integrating the CO emission over the same area, a total flux of 161.6 Jy km s −1 . From this, we estimate that with ALMA, we detect 55.5% of the emission seen in the BIMA observations. Thus a significant fraction of the CO emission in the northern bridge is in an extended component not detected by ALMA. Figure 2. (a)CO (1-0)emission contours (Vollmer et al. 2021)taken with PdBI (beam size 2.7 ×2.7 arcsec 2 )superimposed on an SDSS i-band image of the Taffy system (UGC 12915/4)including the bridge emission. (b)A false-color image of the ALMA total velocity-integrated intensity CO (2-1)(227GHz)mosaic at full resolution (synthesized beam size 0.28 ×0.14 arcsec 2 ), built from 37 separate overlapping primary beam pointings. On this large-scale view, the beam size is too small to display. The arrows show the position of the extragalactic H II region (circle 3″, marked as X-H II)on both figures, and the inset shows the position of the peak in 6 GHz nuclear radio emission (cross; P. N. Appleton et al. 2022, in preparation)in both galaxies superimposed on a zoomed-in image of the CO (2-1) emission. 1″at D=62 Mpc corresponds to 300 pc. Figure 3. (a)Smoothed (to 0.4 ×0.4 arcsec 2 )image of the CO (2-1)total velocity-integrated intensity map showing fainter features. The color stretch encompasses emission as faint as 0.05 Jy km s −1 beam −1 (dark blue)to the brightest at 2 Jy kms s −1 beam −1 (white), and (b)intensity-weighted mean heliocentric velocity (optical velocity definition relative to 4350 km s −1 )of the full field. 7 The Astrophysical Journal, 931:121 (23pp), 2022 June 1 Appleton et al. 3.2. Large-scale Molecular Gas Kinematics In Figure 3(b), we show an intensity-weighted (1st Moment) mean velocity map of the whole system (relative to the average heliocentric velocity of 4350 km s −1 for the two galaxies)with the same 0.4 ×0.4 arcsec 2 smoothing as Figure 3(a). For reference, the systemic velocities of UGC 12915 and UGC 12914 are quite similar (−14 and +21 km s −1 , respectively, implying that most of the radial motion between the galaxies is in the plane of the sky (Condon et al. 1993). In this edge-on view of the collision, the counterrotation of the two galaxies is particularly obvious. UGC 12915 appears to show blueshifted emission in the southeast and redshifted emission in the northwest, whereas UGC 12914 shows the opposite behavior. This counterrotation may have contributed to an increased amount of cloud-cloud collisions when the two galaxies originally collided, almost face-on (Vollmer et al. 2012; Yeager & Struck 2019,2020a,2020b). The extension of scattered gas clouds to the northwest of UGC 12915, discussed earlier, shows peculiar motions not consistent with regular rotation within UGC 12915. While the regular rotation of UGC 12915 is from blueshifted (−340 km s −1 ; blue color)to redshifted gas (+320 km s −1 ; yellow color)in the figure, the clouds farther north and to the west show systemic velocities between −50 and 100 km s −1 (dark green and green). Similarly, noncorotating emission was noted in the velocity field of the ionized gas in this region, where even more discrepant velocities were observed (Joshi et al. 2019). This supports the idea that it may be part of a second, kinematically distinct bridge between the two galaxies, or the remnants of a tidal tail from UGC 12914 (Vollmer et al. 2021). In the main CO bridge, the average kinematics is quite complex, as Figure 4(a)shows (displaying a velocity range from −100 to 140 km s −1 ). The bridge as a whole does not show large-scale coherent motion, but is rather made up of many clumps and filaments with disparate average velocities. Some of the filaments in the cone-shaped structure do show weak systematic motions along parts of their length, but, except for the filaments near UGC 12915, appear kinematically distinct and do not seem obviously related to each other. Figure 4(b)shows a representation of the CO (2-1)line width as a function of position in the bridge. The color coding is based on the FWHM (in km s −1 )for the regions presented in Table A1. The points are superimposed on a contour map of the total intensity of the CO (2-1)emission. Blue and green points show the gas with the lowest line width, whereas many regions have FWHM greater than 65 km s −1 (orange and dark red filled circles), with values extending up to 115 km s −1 . Regions of high velocity dispersion are scattered throughout the filaments and clumps, with the quiescent gas (FWHM <40 km s −1 ) being the minority. Regions with the lowest velocity dispersion are mainly found in some of the filaments close to UGC 12915, and in the area near the X-H II region. Some example spectra are shown in Figure A1. Although it is often traditional to show channel maps (intensity maps of the line emission as a function of velocity)of the full velocity cube of the observations, we will defer this to the second paper, since the emphasis of the current paper is the star formation properties of the bridge. 3.3. Molecular Surface Density in Selected Regions We extract the spectra of more than 239 small regions of the CO emission within the filaments and clumps, and compare their properties to star formation rate estimates derived from the NICMOS observations. In Figure 5, the CO bridge filaments and cloud complexes are divided into 12 large regions, which represent areas that seem spatially and kinematically related. Regions A to G cover emission complexes associated with the brighter part of the northern bridge. Region D includes emission associated with the X-H II region. We also including two coherent filaments, H and I, that fall close to the disk of Figure 4. (a)A zoom-in on the mean velocity field (heliocentric velocity relative to 4350 km s −1 )of the northern bridge to emphasize more finely the velocity structure. White contours show the Paαemission from the X-H II region. (b)A color-coded representation of the fitted FWHM (km s −1 )of the CO-emitting clouds on 85 pc scales superimposed on contours of CO surface density. Green and blue filled circles show clouds low dispersion, whereas orange and red circles show higher values. Only measurements with high-quality profiles (Spectral Quality 1.0, see Table A1)are shown. 8 The Astrophysical Journal, 931:121 (23pp), 2022 June 1 Appleton et al. UGC 12915 and may be part of the bridge. Regions J, to the northwest, and the possible extension of filament A, called AE1, and two bright bridge clumps AE2 are also analyzed. These bridge regions lie outside the area sampled by NICMOS, but are included for completeness. To study the details of the emission in each of the regions shown in Figure 5, we split each structure up into small extraction aperture subregions, which are described more fully in the Appendix, and Figures A2 and A3. Using the software package CASA and the CASA Viewer, we extracted spectra of each region (of dimension 0.23 ×0.28 arcsecs 2 ), slightly larger than the resolution of the ALMA data, and large enough to sample a significant part of the NICMOS PSF. Each of the extracted spectra were then fit with a Gaussian line profile. In a few positions (Regions A1 through A4), we observed doubleline profiles along the same line of sight. Here we fit two Gaussian components. For all of the CO spectra, we estimate the mean radial velocity, the FWHM and peak flux density, and finally the integrated CO flux S CO(2−1) ΔVin Jy km s −1 . The line properties of all of the extracted regions are presented in Table A1. Table A1 provides a flag of the quality of each spectrum. Of all of the spectra extracted, 219 were deemed to be of sufficient quality (good baselines, S/N)to be included in the analysis. To estimate the molecular gas properties from the observed cloud line properties, it is necessary to make several assumptions, including a conversion to H 2 column density. We will use the standard conversion to molecular gas mass, including a 36% correction for helium (Bolatto et al. 2013): ()() () =´ D + --  M MXS VD z1.05 10 1 , 1 L gas 4CO,20 CO 1 0 21 ⎜⎟ ⎛ ⎝⎞ ⎠ where D L 2 is the luminosity distance in megaparsecs, and X CO,20 is the conversion factor N(H 2 )/I CO in units of 2 ×10 20 cm −2 ( ) -- Kkms 11 . Here N(H 2 )is the H 2 molecular column density, and I CO is the velocity-integrated intensity of the CO (1-0) transition in K km s −1 . Therefore, to derive M gas , we also need to make assumptions about both the value of X CO , and the ratio of S CO(1−0) /S CO(2−1) . Braine et al. (2003)suggested that X CO was probably at least a factor of four times lower in the bridge than the Galactic value, based on single-dish observation of the ratio of the 13 CO/ 12 CO (for the 1-0 transition), which implied the 12 CO line was almost optically thin. A similar conclusion was reached by Zhu et al. (2007), using the transitions CO (3-2),(21), and (1-0), and performing Large Velocity Gradient (LVG) modeling (Goldreich & Kwan 1974). They estimated that in the bridge, X CO was between 2–3.6 ×10 19 cm −2 (Kkms −1 ) −1 , which is five to 10 times lower than X CO,20 . Both of these measurements were made with large filled apertures on scales of ∼11″–12″. Recently, Vollmer et al. (2021)adopted an intermediate value of X CO =1/3X CO,20 for the bridge in their PdBI ∼3″resolution beam. Given the uncertainty in the LVG modeling, and the large difference in scale between the previous single-dish observations and our ALMA observations, we adopt as an initial working hypothesis the Braine et al. (2003)value of X CO =5×10 19 cm −2 (Kkms −1 ) −1 ,or1/4 X CO,20 . We will explore the implications of varying this value on the derived properties of the clouds and their line-of-sight extinction under different assumptions. We finally assume ()() () () nn= ---- - S Sr CO10 CO21 10 21 221 1 , and r 21 =0.79 (see Zhu et al. 2007). Gas masses from our extracted regions range from 0.1–1×10 7 (X CO /X CO,20 )M e . 4. Ionized Gas Emission in the Bridge and X-H II Region 4.1. The PaαDistribution in the Bridge In Figure 6we show the complex filamentary structure of part of the bridge in CO (2-1)superimposed on the Paα NICMOS image of UGC 12915, which includes the brighter and most interesting parts of the bridge region. The figure shows that within UGC 12915, the CO is largely confined within the inner disk, and is well correlated with the bright Paα emission. A narrow nuclear CO disk is detected, and there are several CO filaments extending away from the disk to the north of UGC 12915, which, except for a few isolated cases, are devoid of obvious star formation. Concentrating on the bridge region, the most prominent Paα emission comes from the bright extended X-H II region (Bushouse 1987; Jarrett et al. 1999; Joshi et al. 2019). Several clumps of CO emission are seen projected against this SF region, including an elongated finger of faint CO emission, which crosses its center. Very little Paαemission is seen from the other CO structures, except for one or two faint possible associations. The lack of obvious star formation in these dense clumps is a characteristic of the CO emission observed in the bridge (see also Vollmer et al. 2021). For those regions with extracted CO spectra that fall within the area covered by NICMOS, we then proceeded to extract Figure 5. The CO (2-1)integrated bridge emission divided into regions of bright emission. Regions A–G include the filaments and clumps associated with the main bridge (where D includes emission associated with the X-H II region)and H and I show filaments that may be part of the bridge but lie close to UGC 12915. Regions J and AE1/2 are scattered regions in the bridge. AE1 may be an extension of filament A. Each region is divided further into many small extracted apertures, the properties of which are shown both graphically and in tabular form in the Appendix (Figures A2 and A3, and Table A1). 9 The Astrophysical Journal, 931:121 (23pp), 2022 June 1 Appleton et al. respectively, based on Paα. The radio continuum provides approximately similar values. Over the next 10 Myr, 10 6 M e of stars could form, making them future young massive star cluster candidates. 6. Other Potentially Similar Environments to the Taffy Bridge Gas 6.1. Comparison with Antennae “Overlap”Region It is worthwhile contrasting the properties of the Taffy bridge with another well-studied colliding galaxy system, the Antennae galaxy pair, NGC4038/9. The Antennae is known to be in the starburst phase of a major merger (Whitmore et al. 1999; Schweizer et al. 2008). Unlike the Taffy system, which has experienced a head-on counterrotating collision in the past (25–35 Myr ago), the Antennae is still in the compressive “contact”stage of a relatively slow (100 Myr)prograde diskdisk merger (Renaud et al. 2018). In global properties, the Antennae and Taffy have very similar total FIR luminosities and total molecular gas masses. The Antennae has an FIR luminosity L FIR =5.6 ×10 10 L e (Brandl et al. 2009)and total molecular mass 1.1 ×10 10 M e (Zhu et al. 2003), whereas equivalent values for the Taffy are L FIR =6.5 ×10 10 L e (Sanders et al. 2003)and a total molecular mass M mol = 0.97 ×10 10 M e (Zhu et al. 2007). The “overlap”region of the Antennae (which lies between the two component galaxies)also shares a number of similarities with the Taffy bridge. They both include significant quantities of molecular gas distributed in narrow filaments. Bearing in mind the uncertainties in the X CO factor for both galaxies, the Taffy bridge has roughly twice as much molecular gas as the “overlap”region. The Taffy bridge contains ∼1.3 ×10 9 M e (for X CO =1/5X CO,20 ; Zhu et al. 2007), compared with ∼0.5 ×10 9 M e for the Antennae “overlap”region (Stanford et al. 1990; Wilson et al. 2000),ifwe assume a common X CO factor. Unlike the Taffy bridge, which we have shown is largely devoid of star formation, the Antennae overlap region is bursting with activity. The total star formation rate in the overlap regions has been measured to be 5 M e yr −1 (Stanford et al. 1990)compared with 0.1–0.25 M e yr −1 for the Taffy bridge (this paper). In the Antennae overlap region, the star formation activity is concentrated in a handful of very massive SGMCs (Wilson et al. 2000). With the one exception of the “Firecracker”discussed earlier, the majority of these giant molecular complexes in the Antennae dominate the star formation output of the overlap region (Mirabel et al. 1998; Whitmore & Schweizer 1995; Whitmore et al. 2010; Johnson et al. 2015). In the Taffy, the molecular clouds are less massive, and only the X-H II region compares in molecular mass (10 8 M e )with the lower end of the mass distribution of the Antennae SGMCs. Another difference is that, unlike the Taffy filaments, which contain clouds with locally high velocity dispersion (40–100 km s −1 ), the Antennae CO filaments exhibit unusually low velocity dispersion (10 km s −1 ; Whitmore et al. 2014). Higher CO velocity dispersion is reported for the Antennae system in general (Leroy et al. 2016), which likely correlates with increased star formation activity. This seems the opposite of the Taffy, where the X-H II region exhibits the lowest velocity dispersion in the CO clouds, whereas the filaments and star formation-deficient clouds have the highest velocity dispersion. In summary, except for some regions of high star formation rate in the Taffy X-H II region, the filaments and clumps in the Taffy bridge are strongly suppressed in star formation compared with the Antennae. 6.2. Comparison with Molecular Gas in Ram-pressure-stripped Systems Another kind of environment that might have some similarities with the Taffy bridge consists of those involving the ram pressure stripping of gas from cluster galaxies, where stripped molecular clouds find themselves in the intergalactic medium. There is now significant evidence that the interstellar media (ISMs)of late-type cluster galaxies can be stripped (Gunn & Gott 1972)by their interaction with a hot gaseous cluster halo, including the formation of tails revealed in HI, Hα emission, X-rays, and warm and cold molecular hydrogen (e.g., Gavazzi et al. 2001; Wang et al. 2004; Oosterloo & van Gorkom 2005; Sun et al. 2007; Yagi et al. 2007; Jáchym et al. 2014). Although the physical mechanism from stripping gas into the intergalactic medium (see Schulz & Struck 2001)is rather different from the case of a major head-on collision, as in the Taffy case, there are some potential similarities. For example, Sivanandam et al. (2010)found evidence for warm molecular hydrogen in the stripped tail of ESO 137–001 that was likely shock-heated by turbulent interactions with its host ICM of Abell cluster 3627. Cold molecular gas has also been found in some likely ram-pressure-stripped galaxies, primarily through single-dish observations of low spatial resolution (e.g., Jáchym et al. 2014,2017; Lee et al. 2017). Only recently have high-resolution observations become available (Jáchym et al. 2019)including those studied by the GASP survey. 22 In these cases, these intergalactic clouds, either stripped from their host galaxies or formed in situ through turbulent compression, can be compared with the Taffy bridge. One particular gas-stripped system, JW 100, is a member of a class of similar “Jellyfish”galaxies (Ebeling et al. 2014), and was observed with ALMA in the CO (1-0)and (2-1)transitions (Moretti et al. 2018,2020). This massive cluster galaxy (with a diameter of at least 45 kpc)exhibits a clumpy tail of molecular hydrogen extending 35 kpc away from its disk. Approximately 7×10 9 M e of molecular gas is present in the tail, which is comparable with the Taffy bridge, representing about 30% of the gas in the galaxy disk. Like the Taffy, several clumps in the tail exhibit high velocity dispersion (70–80 km s −1 ), although much of the gas exhibits lower linewidths. Although the mechanism for stripping the gas in such tails may be markedly different from the Taffy (Pedrini et al. 2022), it is possible that gas may be formed in shocks or eddies in the wake of the main shocked region around the galaxy. Is there evidence for star formation suppression here too? In Figure 13 we show examples taken from Moretti et al. (2020)of clouds in the stripped tail of the JW100 (blue circles) overplotted on a partially grayed-out version of one characteristic KS relationship plot of the Taffy from this paper (Figure 10(b)). As is discussed in more detail by Moretti et al. (2020), many of the stripped clouds fall below the standard KS relationship. Thus, like the majority of nondetected Paαupper limits in Taffy, the stripped clouds seem to show very low star formation rates, as measured by optical spectroscopy. One difference between those stripped clouds and Taffy is that the former appear to have significantly lower 22 GASP =Gas Stripping Phenomena with MUSE (Poggianti et al. 2017). 16 The Astrophysical Journal, 931:121 (23pp), 2022 June 1 Appleton et al. overall H 2 surface densities compared with the Taffy bridge, even if (as was assumed by the authors)X CO has a galactic value. However, one should caution that the ALMA data discussed by Moretti et al. (2020)was taken with 1″resolution, which corresponds to 1 kpc at the distance of JW100. Thus the surface densities of molecular gas could well be much higher than observed, if the system had been observed at the same linear resolution as the Taffy. Nevertheless, it is clear that there are some similarities between the properties of the gas in the stripped clouds in the JW 100 and those suppressed star formation clouds in the Taffy bridge. The careful comparison of the multiwavelength observations of JW100 by Poggianti et al. (2019)further strengthens those similarities. Like the Taffy X-H II region, a ULX source was found to be associated with one of the brighter H II regions. These authors also showed that although parts of the stripped gas show evidence of star formation, there are regions of the tail that show LINER-type optical emission (similar to regions of the Taffy bridge; Joshi et al. 2019), and X-ray emission not consistent with star formation. A variety of possible heating mechanisms for such systems, including strong turbulent mixing, plasma interactions, thermal conduction, and shocks, have been put forward as possible explanations (Poggianti et al. 2019; Campitiello et al. 2021; Pedrini et al. 2022). 7. Origin of the Star Formation Suppression in the Bridge The majority of the clouds in the Taffy bridge are likely unbound if self-gravity is the only restoring force for a wide range of reasonable values of assumed X CO . Such clouds would evaporate on a crossing time (2R/σ), which is typically 2–8 Myr (see Figure 14)for the average Taffy bridge cloud. This is much shorter then the nominal time since the Taffy galaxies collided of 25–35 Myr (Condon et al. 1993). This short timescale seems to rule out the possibility that the clouds in the bridge were originally normal molecular clouds confined to their host disks before the collision. In such a simple picture, such clouds would suddenly no longer feel the gravitational forces of the original disk, and may be expected to freely expand when ejected into intergalactic space. Although it is possible that a subset of clouds (the long tail of the distribution in Figure 14)may have crossing times long enough to have survived without an external medium since the collision (we cannot resolve the linewidths of clouds with FWHM < 15 km s −1 ), it is clear that the majority of the gas, including those associated with the X-H II region, have crossing times that suggest they have formed relatively recently. Furthermore, models of head-on collisions like the Taffy system show significant ionization of the ISMs in both galaxies at and shortly after the collision, suggesting that only a relatively small number of clouds punch through unscathed (Yeager & Struck 2020a,2020b). To understand why the CO clouds are not virialized, we are reminded that the Taffy bridge CO clouds are embedded in a highly turbulent multiphase medium. There is evidence for the existence of copious quantities of shock-heated warm H 2 through direct mid-IR emission lines, and indirectly through the detection of [CI]and [CII]emission (Peterson et al. 2012,2018). Additionally, there is evidence for shock-excited atomic emission-line gas in the bridge (Joshi et al. 2019). Thus the CO-emitting clouds are likely immersed in a highly turbulent medium, similar to the intergalactic shocked filament in Stephan’s Quintet (Appleton et al. 2017; Guillard et al. 2022). Vollmer et al. (2021)discussed this in the context of their dynamical model of the Taffy system, which was tuned to provide a good match with the lower-resolution PdBI CO observations of the Taffy. In that paper, the authors use their model to show that turbulence driven by cloud-cloud collisions on large scales in the bridge would lead to clouds that are not virialized. Our observations, which increase the spatial resolution by an order of magnitude compared with the PdBI observations, fully support this conclusion. However, the actual details of the mechanism by which cold CO clouds are dynamically heated are likely to be quite complex. For example, the cascade of energy from the large driving scales of the collision, and subsequent continued high Mach-number cloud-cloud collisions can only be properly modeled with the multiphase model, which takes into account atomic and molecular cooling phases, phase transitions, shocks, turbulence and gas phase-mixing in different media (Guillard et al. 2009). For example, in the analogous Stephan’s Quintet shocked filament, recent observations of Lyαemission Figure 13. A comparison between the surface formation rate density and molecular hydrogen surface density for clouds in the ram-pressure-stripped tail of the jellyfish galaxy JW100 (blue filled circles, where X CO was assumed to be Galactic; Moretti et al. 2020), and a partially grayed-out version of Figure 10(b) of the current paper (the symbols are the same as in Figure 10). Like the majority of the Taffy clouds, which have upper limits to star formation, the JW100 clouds also show significant star formation suppression. Figure 14. Cloud crossing times in the bridge in megayears for an assumed diameter of 85 pc. If the clouds are smaller, the crossing times will be even shorter. The clouds devoid of star formation (solid blue)have generally shorter crossing times than those containing star formation (solid red). The other samples are the same as those described in Figure 9. 17 The Astrophysical Journal, 931:121 (23pp), 2022 June 1 Appleton et al. (Guillard et al. 2022)combined with previous observations of molecular hydrogen and X-rays (Appleton et al. 2017; O’Sullivan et al. 2009)show strong evidence that the dissipation of kinetic energy involves all gas phases. These are likely connected through a turbulent cascade involving atomic and molecular magnetohydrodynamic shocks (Lesaffre et al. 2020; Durrive et al. 2021), and turbulent mixing layers (Slavin et al. 1993). The models of Yeager & Struck (2020a,2020b)are a step in the right direction in attempting to treat multiphase cooling of the gas in Taffy-like bridge systems. Those models predict rapid evolution in the gas phases as the collision progresses as shocks permeate the medium, with the appearance of waves of cooling, and periods of re-excitation by cloud-cloud collisions. The inclusion of mid-IR H 2 and [CII]cooling lines is another important step (T. Yeager et al. 2022, in preparation). These models, however, cannot yet capture the turbulent mixing that must occur at the boundaries between the hot, warm, and cold components. Such models only partially deal with the possibility that the cold molecular clouds are transient material that is constantly being recycled through different thermal phases. In such cases, clouds must cool and collapse rapidly to have a chance of forming stars. The existence of high Mach-number gas collisions may stochastically generate regions like the X-H II region (Yeager & Struck 2020b). The Paαhotspots in the X-H II region could be examples of the formation of massive proto-clusters conceptually similar to fast gas-on-gas collisions between dwarf galaxies postulated as one formation mechanism for ultradiffuse galaxies (Silk 2019). The hotspots have locally high star formation surface densities, and over the next 10 Myr, the extragalactic H II region could form ∼1.3 ×10 6 M e of stars, making them future young massive star cluster candidates. Although in the case of the Taffy, where the size of the colliding systems is larger than Silk (2019)envisaged, high Mach-number gas collisions will likely continue to generate regions like the X-H II region across the full extent of the bridge over time (Yeager & Struck 2020b). An added complexity is the possible influence of magnetic fields. The Taffy was discovered because of the detection of the radio continuum between the galaxies resulting from synchrotron radiation from cosmic rays spiraling in magnetic fields (Condon et al. 1993). Lisenfeld & Völk (2010)explained the radio emission as the result of cosmic rays accelerated in shocks produced in situ in the Taffy bridge. The powerful emission from the lowest pure-rotational transitions of H 2 in Stephan’s Quintet, which exhibits very similar Mid-IR H 2 spectra to the Taffy bridge, is most easily explained (Lesaffre et al. 2013)by magnetic shocks, which moderate the temperature of the shock into the temperature range where the 0-0S(0)and 0-0S(1)lines dominate (100 <T<300 K). New deep VLA radio continuum observations with much higher resolution than the original Condon et al. (1993) observation have been recently made (P. N. Appleton 2022, in preparation), which will provide a more complete picture of the distribution of thermal and nonthermal emission in the Taffy bridge, and may help determine the degree of importance of magnetic fields in the process of energy dissipation and cloud dynamics. 8. Conclusions This is the first of two papers describing ALMA observations of CO in the Taffy galaxy system, UGC 12914/5. In this paper we present a large mosaic of primary beam pointings made in the CO (2-1)transition at a spatial resolution more than an order of magnitude greater than previous observations (0.24 ×0.18 arcsec 2 ). The paper concentrates on the molecular bridge extending between the galaxies. When compared with archival HST NICMOS Paαand Spitzer observations, as well as new VLA radio continuum, Palomar 5 m optical spectroscopy and previously published Chandra X-ray observations, we conclude the following: 1. The observations provide evidence that gas on scales of 60–100 pc in the Taffy galaxies is significantly disturbed in the aftermath of the head-on collision between the galaxies. The gas in UGC 12915 is particularly disturbed, with streamers of likely tidal debris extending northward, far from the inner disk. UGC 12914, although less obviously disturbed that its companion, exhibits a singular powerful narrow structure of CO, which follows the southern edge of the inner disk into a hook-like arm. 2. The molecular gas bridge between UGC 12914/5is resolved into myriad narrow filaments (60–100 pc ×1 kpc) and clumpy structures extending between the two galaxies that have some similarities with the filaments seen in the “overlap”region of the Antennae galaxies, except that almost all of them are devoid of star formation. By analyzing 234 regions in the bridge, we compare the star formation rate surface density derived for PaαHST observations with the ALMA data at 85 pc resolution. We show that the majority of the filaments are devoid of star formation, and fall significantly below the Kennicutt– Schmidt relationship for normal galaxies, especially for the numerous regions undetected in Paα. This statement is shown to be true over a wide range of assumed extinction, X CO values, and dust-to-gas ratios. The result strongly supports the idea that, except in the X-H II region, the gas in the bridge is experiencing strong star formation suppression, even at the high surface densities probed by the new ALMA observations. 3. The kinematics of the majority of the bridge molecular gas shows unusually high velocity dispersion on the scale of 85 pc, with some regions within the filaments showing Gaussian line profiles with FWHM 100 km s −1 .Gas associated with the X-H II region tends to have lower velocity dispersion, suggesting it is less turbulent there. Like clouds seen in the Antennae galaxies, the Taffy clouds show a steeper slope in log(σ)versus log(Σ gas )than normal galaxies. In addition, most of the Taffy bridge regions are gravitationally unbound for most reasonable values of assumed X CO (1/4<(X CO /X CO,20 )<1/2), and, like the “Firecracker”region in the Antennae system, would require an extremely high external pressure (10 6 <P e /k<10 8 )to pressure-support them. If they were not externally supported, many of the clouds would disperse on a crossing timescale of ∼2–5 Myr and would therefore likely be transient. Such clouds may be continuously created and destroyed in such a highly turbulent medium. 4. Despite the highly turbulent medium, stars are able to sometimes form there, albeit at a low rate. In the X-H II region, we derive a star formation rate of 0.1–0.25 M e yr −1 . 18 The Astrophysical Journal, 931:121 (23pp), 2022 June 1 Appleton et al. Gas associated with this region exhibits high local star formation rates, on or above the KS relationship for normal systems depending on the assumed X CO value. These higher star formation rate densities of 0.6–0.8 M e yr −1 kpc −2 in two regions (containing faint compact radio sources and a ULX source)on scales of <100 pc might be evidence of the formation of massive star clusters. We also present evidence of potential rotational motions in the molecular gas and ionized gas associated with the X-H II region. The discovery that the majority of the dense molecular clouds in the bridge are not able to form stars suggests that turbulence, driven by the collision of the two Taffy galaxies more than 25–30 Myr previously, is still actively suppressing star formation down to the scale of 60–100 pc across the majority of the bridge. Recent gas-dynamical models of Taffylike head-on collisions support the idea that even after 30 Myr, cloud-cloud collisions are still creating high Mach-number collisions in the bridge. Such collisions may also explain the large quantities of warm molecular hydrogen discovered in the bridge by the Spitzer IRS, as well as the existence of extensive shock-excited ionized gas throughout the bridge, including strong Hαemission at a discrepant velocity near the X-H II region in the bridge. The process in which large-scale-driven turbulence caused by the galaxy collision can influence dense cold molecular clouds on 60–100 pc scales is likely to involve many different processes, including atomic and molecular magnetic shocks and turbulent mixing of gas, similar to that suspected in the Stephan’s Quintet system. This paper made use of data from the following observatories/instruments: the Atacama Large Millimeter/submillimeter Array (ALMA), the Hubble Space Telescope (HST; NICMOS), the Chandra X-ray telescope S3 Advanced CCD Imaging Spectrometer, the Very Large Array (VLA), the Spitzer Space telescope, as well as IRAC and MIPS and the Double Beam Spectrograph on the 5 m Hale telescope at the Palomar Observatory. The paper made use of the following software: Community Astronomy Software Applications (CASA; NRAO), IDL (L3Harris Geospatial Solutions), Python 3 open source software, PypeIt (Prochaska et al. 2020), and SAOImage DS9 (Chandra X-ray Science Center, HEASARC and JWST Mission office at STSCI.) This paper makes use of the following ALMA data: ADS/ JAO.ALMA.2016.1.01037.S. ALMA is a partnership of ESO (representing its member states), NSF (USA), and NINS (Japan), together with NRC (Canada), MOST and ASIAA (Taiwan), and KASI (Republic of Korea), in cooperation with the Republic of Chile. The Joint ALMA Observatory is operated by ESO, AUI/NRAO, and NAOJ. The DBSP spectra are based on observations obtained at the Hale Telescope, Palomar Observatory as part of a continuing collaboration between California Institute of Technology, NASA/JPL, Yale University, and the National Astronomical Observatories of China. This work also contains archival data obtained with the Spitzer Space Telescope, which was operated by the Jet Propulsion Laboratory, California Institute of Technology under a contract with NASA. The National Radio Astronomy Observatory is a facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc. This research has made use of the NASA/IPAC Infrared Science Archive, which is funded by the National Aeronautics and Space Administration and operated by the California Institute of Technology. P.A. would like to acknowledge the NRAO visitor support during a visit to NRAO-Charlottesville early in this project. P.A. would also like to thank the Institut d’Astrophysique de Paris for support as a Visiting Scientist in 2019 October in connection with this work. U.L. acknowledges support from project PID2020-114414GB-100, financed by MCIN/AEI/10.13039/ 501100011033, from projects P20_00334 and FQM108, financed by the Junta de Andalucia and from FEDER/Junta de Andalucía-Consejería de Transformación Económica, Industria, Conocimiento y Universidades/Proyecto A-FQM510-UGR20. The authors wish to thank the anonymous referee for many important suggestions that improved this paper. Appendix Extracted Regions In this Appendix we provide a sample table (in machinereadable format)of the regions extracted in the bridge and the measured and derived properties. We also present several figures which provide some additional information about the extraction regions. This includes Figure A1, which provides some illustrative CO spectra of selected regions. In Table 1, Column 1 provides the region that corresponds to the regions marked in Figures A2 and A3, and were chosen to be representative regions in the bridge structures. Column 2 and 3 are the R.A. (J2000)and decl. (J2000)of the regions. Column 4 and 5 are the best-fitting systemic heliocentric velocity (optical definition)and FWHM, with uncertainties, of the CO profiles after fitting with a Gaussian in km s −1 . Column 6 is the integrated flux of the CO (2-1)profile in Jy km s −1 and uncertainties. Column 7 is a spectral quality flag based. Flags of unity are the best quality spectra, and the table notes explain the other flags. Only Quality 1 flagged data are used in the figures presented in the paper. Column 8 provides the molecular mass surface density, in units of log 10 (M e pc −2 ), for the nominal case of X CO =1/4X CO,20 . In this paper, we consider potentially realistic cases of X CO in the range 1/4<X CO,20 <1/2, where XCO,20 is the standard Galactic value. Column 9 tabulates the surface density of Paαemission in units of log 10 (erg s −1 cm −2 kpc −2 )for the Case 1 extinction example given in Section 5.1. Column 10 gives the star formation surface density in units of log 10 (M e yr −1 kpc −2 ) under the same assumptions. Column 11 provides a Paα detection flag (see table note). 19 The Astrophysical Journal, 931:121 (23pp), 2022 June 1 Appleton et al. Figure A1. Examples of spectra described in Table A1 and in Figures A2 and A3. The designation refers to the entry in the Table, and the blue lines show Gaussian fits to the spectra. Narrow lines are seen in the region associated with the X-H II region (e.g., D9 and D12). 20 The Astrophysical Journal, 931:121 (23pp), 2022 June 1 Appleton et al. Figure A2. Extraction apertures for regions A, B, C, D, and E superimposed on the integrated CO surface density image. Figure A3. Extraction apertures for regions F, G, H, I, J, AE1, and AE2 superimposed on the integrated CO surface density image. 21 The Astrophysical Journal, 931:121 (23pp), 2022 June 1 Appleton et al. ORCID iDs P. N. Appleton https://orcid.org/0000-0002-7607-8766 B. Emonts https://orcid.org/0000-0003-2983-815X U. Lisenfeld https://orcid.org/0000-0002-9471-5423 E. Falgarone https://orcid.org/0000-0003-0693-2477 P. Guillard https://orcid.org/0000-0002-2421-1350 F. Boulanger https://orcid.org/0000-0003-1097-6042 P. Ogle https://orcid.org/0000-0002-3471-981X C. Struck https://orcid.org/0000-0002-6490-2156 B. Vollmer https://orcid.org/0000-0003-1740-1284 T. Yeager https://orcid.org/0000-0002-2582-0190 References Anselmet, F., Antonia, R. A., & Danaila, L. 2001, P&SS,49, 1177 Appleton, P. N., Charmandaris, V., & Struck, C. 1996, ApJ,468, 532 Appleton, P. N., Guillard, P., Togi, A., et al. 2017, ApJ,836, 76 Appleton, P. N., Lanz, L., Bitsakis, T., et al. 2015, ApJ,812, 118 Appleton, P. N., & Struck-Marcell, C. 1987, ApJ,318, 103 Appleton, P. N., Xu, K. C., Reach, W., et al. 2006, ApJ,639, 51 Armus, L., Charmandaris, V., & Soifer, B. T. 2020, NatAs,4, 467 Armus, L., Heckman, T., & Miley, G. 1987, AJ,94, 831 Armus, L., Mazzarella, J. M., Evans, A. S., et al. 2009, PASP,121, 559 Bohlin, R. C., Savage, B. D., & Drake, J. F. 1978, ApJ,224, 132 Bolatto, A. D., Wolfire, M., & Leroy, A. K. 2013, ARA&A,51, 207 Braine, J., Davoust, E., Zhu, M., et al. 2003, A&A,408, L13 Braine, J., Lisenfeld, U., Duc, P.-A., et al. 2004, A&A,418, 419 Brandl, B. R., Snijders, L., den Brok, M., et al. 2009, ApJ,699, 1982 Bushouse, H. A. 1987, ApJ,320, 49 Calzetti, D., Kennicutt, R. C., Engelbracht, C. W., et al. 2007, ApJ,666, 870 Calzetti, D., Kinney, A. L., & Storchi-Bergmann, T. 1994, ApJ,429, 582 Calzetti, D., Wu, S.-Y., Hong, S., et al. 2010, ApJ,714, 1256 Campitiello, M. G., Ignesti, A., Gitti, M., et al. 2021, ApJ,911, 144 Charmandaris, V., & Appleton, P. N. 1996, ApJ,460, 686 Cluver, M. E., Appleton, P. N., Boulanger, F., et al. 2010, ApJ,710, 248 Condon, J. J. 1992, ARA&A,30, 575 Condon, J. J., Helou, G., Sanders, D. B., et al. 1993, AJ,105, 1730 Cornwell, T. J. 2008, ISTSP,2, 793 de los Reyes, M. A. C., & Kennicutt, R. C. 2019, ApJ,872, 16 Durrive, J.-B., Ferrière, K., & Lesaffre, P. 2021, arXiv:2101.03447 Ebeling, H., Stephenson, L. N., & Edge, A. C. 2014, ApJL,781, L40 Fazio, G. G., Hora, J. L., Allen, L. E., et al. 2004, ApJS,154, 10 Field, G. B., Blackman, E. G., & Keto, E. R. 2011, MNRAS,416, 710 Gao, Y., Zhu, M., & Seaquist, E. R. 2003, AJ,126, 2171 Gavazzi, G., Boselli, A., Mayer, L., et al. 2001, ApJL,563, L23 Gerber, R. A., Lamb, S. A., & Balsara, D. S. 1996, MNRAS,278, 345 Gordon, K. D., Clayton, G. C., Misselt, K. A., et al. 2003, ApJ,594, 279 Guillard, P., Appleton, P. N., Boulanger, F., et al. 2022, ApJ,925, 63 Guillard, P., Boulanger, F., Pineau Des Forêts, G., et al. 2009, A&A,502, 515 Gunn, J. E., & Gott, J. R. 1972, ApJ,176, 1 Haffner, L. M., Dettmar, R.-J., Beckman, J. E., et al. 2009, RvMP,81, 969 Higdon, J. L. 1996, ApJ,467, 241 Hily-Blant, P., & Falgarone, E. 2009, A&A,500, L29 Högbom, J. A. 1974, A&AS, 15, 417 Jáchym, P., Combes, F., Cortese, L., et al. 2014, ApJ,792, 11 Jáchym, P., Kenney, J. D. P., Sun, M., et al. 2019, ApJ,883, 145 Jáchym, P., Sun, M., Kenney, J. D. P., et al. 2017, ApJ,839, 114 Jarrett, T. H., Helou, G., Van Buren, D., et al. 1999, AJ,118, 2132 Johnson, K. E., Leroy, A. K., Indebetouw, R., et al. 2015, ApJ,806, 35 Jones, A. P., Tielens, A. G. G. M., & Hollenbach, D. J. 1996, ApJ,469, 740 Joshi, B. A., Appleton, P. N., Blanc, G. A., et al. 2019, ApJ,878, 161 Joye, W. A., & Mandel, E. 2003, in Astronomical Data Analysis Software and Systems XII, ASP Conf. Ser. 295, ed. H. E. Payne et al. (San Francisco, CA: ASP),489 Kahre, L., Walterbos, R. A., Kim, H., et al. 2018, ApJ,855, 133 Kennicutt, R. C. 1998a, ApJ,498, 541 Kennicutt, R. C. 1998b, ARA&A,36, 189 Kennicutt, R. C., Calzetti, D., Walter, F., et al. 2007, ApJ,671, 333 Konstantopoulos, I. S., Appleton, P. N., Guillard, P., et al. 2014, ApJ,784, 1 Lee, B., Chung, A., Tonnesen, S., et al. 2017, MNRAS,466, 1382 Leroy, A. K., Hughes, A., Schruba, A., et al. 2016, ApJ,831, 16 Leroy, A. K., Schinnerer, E., Hughes, A., et al. 2017, ApJ,846, 71 Leroy, A. K., Schinnerer, E., Hughes, A., et al. 2018, VizieR On-line Data Catalog: J/ApJ/846/71 Lesaffre, P., Pineau des Forêts, G., Godard, B., et al. 2013, A&A,550, A106 Lesaffre, P., Todorov, P., Levrier, F., et al. 2020, MNRAS,495, 816 Lisenfeld, U., & Völk, H. J. 2010, A&A,524, A27 Lynds, R., & Toomre, A. 1976, ApJ,209, 382 Madau, P., Lupi, A., Diemand, J., et al. 2020, ApJ,890, 18 Meidt, S. E., Glover, S. C. O., Kruijssen, J. M. D., et al. 2020, ApJ,892, 73 Meidt, S. E., Leroy, A. K., Rosolowsky, E., et al. 2018, ApJ,854, 100 Mihos, J. C., & Hernquist, L. 1996, ApJ,464, 641 Mirabel, I. F., Vigroux, L., Charmandaris, V., et al. 1998, A&A, 333, L1 Moretti, A., Paladino, R., Poggianti, B. M., et al. 2018, MNRAS,480, 2508 Table A1 Properties of the Extracted Regions Reg. R.A. Decl. V sys b FWHM c ΣS V ΔVSpec d Log(Σ mol ) e Σ(Paα) f Log(Σ SFR ) f Paα g #(J2000)(J2000)CO (2-1)CO (2-1)CO (2-1)Qual. 1/4X CO,20 (10 −16 ×erg s −1 (M e yr −1 Det. (hh:mm:ss)(dd:mm:ss)(km s −1 )(km s −1 )(Jy km s −1 )(M e pc −2 )-- cm arcsec 22 )kpc −2 )Flag [1][2][3][4][5][6][7][8]([9][10][11] A1a 00:01:41.479 +23:29:37.1 4380 96 ±14 0.29 ±0.06 1.0 2.46 ±0.08 a 0.43 a −1.75 a 0 A1b 00:01:41.479 +23:29:37.1 4510 28 ±4 0.08 ±0.02 1.0 0 A2a 00:01:41.466 +23:29:37.0 4376 32 ±5 0.09 ±0.02 1.0 2.27 ±0.08 a 0.43 a −1.75 a 0 A2b 00:01:41.466 +23:29:37.0 4519 28 ±3 0.14 ±0.02 1.0 0 A3a 00:01:41.463 +23:29:36.7 4375 36 ±7 0.09 ±0.02 1.0 2.23 ±0.11 a 0.43 a −1.75 a 0 A3b 00:01:41.463 +23:29:36.7 4522 30 ±3 0.13 ±0.01 1.0 0 A4a 00:01:41.448 +23:29:36.4 4376 41 ±6 0.11 ±0.02 1.0 2.33 ±0.08 a 0.43 a −1.75 a 0 A4b 00:01:41.448 +23:29:36.4 4508 47 ±3 0.16 ±0.02 1.0 0 Notes. a Regions A1–A4 have multiple CO components 1a, 1b, 2a, 2b, etc. Their integrated Paαflux and SFR properties are combined. b Heliocentric radial velocity using the optical definition. c FWHM corrected for instrumental resolution. d Quality factor of CO spectrum, 1 =good, 0.7 =weak detection, 0.5 =very weak, 0.4 =poor, 0 =undetected in CO line. Only quality 1 are shown in the figures. e Molecular surface density for nominal case of X CO =1/4X CO,20 ; see the text. f Surface flux and star formation rates estimated from the Paαemission assuming minimal (Case 1)extinction (see Section 5.1). g Paαdetection flags; Flag =0 indicates an upper limit, Flag =1 indicates a detection, and Flag =−1 indicates that the region lies outside the PaαNICMOS coverage area. Note. Table A1 is published in its entirety in machine-readable format. A portion is shown here for guidance regarding its form and content. (This table is available in its entirety in machine-readable form.) 22 The Astrophysical Journal, 931:121 (23pp), 2022 June 1 Appleton et al. Moretti, A., Paladino, R., Poggianti, B. M., et al. 2020, ApJ,889, 9 Murphy, E. J., Condon, J. J., Schinnerer, E., et al. 2011, ApJ,737, 67 O’Sullivan, E., Giacintucci, S., Vrtilek, J. M., et al. 2009, ApJ,701, 1560 Oosterloo, T., & van Gorkom, J. 2005, A&A,437, L19 Osterbrock, D. E. 1989, in Astrophysics of Gaseous Nebulae and Active Galactic Nuclei, ed. D. E. Osterbrock (Mill Valley, CA: University Science Books),408 Padoan, P., Haugbølle, T., & Nordlund, Å2012, ApJL,759, L27 Pedrini, A., Fossati, M., Gavazzi, G., et al. 2022, MNRAS,511, 5180 Peterson, B. W., Appleton, P. N., Bitsakis, T., et al. 2018, ApJ,855, 141 Peterson, B. W., Appleton, P. N., Helou, G., et al. 2012, ApJ,751, 11 Poggianti, B. M., Ignesti, A., Gitti, M., et al. 2019, ApJ,887, 155 Poggianti, B. M., Moretti, A., Gullieuszik, M., et al. 2017, ApJ,844, 48 Prochaska, J., Hennawi, J., Westfall, K., et al. 2020, JOSS,5, 2308 Rahmer, G., Smith, R. M., Bui, K., et al. 2012, Proc. SPIE,8446, 84462Z Relaño, M., Lisenfeld, U., Pérez-González, P. G., et al. 2007, ApJL,667, L141 Renaud, F., Athanassoula, E., Amram, P., et al. 2018, MNRAS,473, 585 Rich, J. W., de Blok, W. J. G., Cornwell, T. J., et al. 2008, AJ,136, 2897 Rieke, G. H., Young, E. T., Engelbracht, C. W., et al. 2004, ApJS,154, 25 Saito, T., Iono, D., Yun, M. S., et al. 2015, ApJ,803, 60 Sanders, D. B., Mazzarella, J. M., Kim, D.-C., Surace, J. A., & Soifer, B. T. 2003, AJ,126, 1607 Sanders, D. B., & Mirabel, I. F. 1996, ARA&A,34, 749 Sanders, D. B., Scoville, N. Z., Young, J. S., et al. 1986, ApJL,305, L45 Schulz, S., & Struck, C. 2001, MNRAS,328, 185 Schweizer, F., Burns, C. R., Madore, B. F., et al. 2008, AJ,136, 1482 Silk, J. 2019, MNRAS,488, L24 Sivanandam, S., Rieke, M. J., & Rieke, G. H. 2010, ApJ,717, 147 Slavin, J. D., Shull, J. M., & Begelman, M. C. 1993, ApJ,407, 83 Soifer, B. T., Helou, G., Lonsdale, C. J., et al. 1984, ApJL,283, L1 Soifer, B. T., Helou, G., Lonsdale, C. J., et al. 1984, ApJL,283, L1 Stanford, S. A., Sargent, A. I., Sanders, D. B., et al. 1990, ApJ,349, 492 Struck, C. 1997, ApJS,113, 269 Struck-Marcell, C. 1990, AJ,99, 71 Sun, M., Donahue, M., & Voit, G. M. 2007, ApJ,671, 190 Toomre, A. 1978, IAU Symp. No. 79, Large Scale Structures in the Universe, Vol. 79 ed. M. S. Longair & J. Einasto (Dordrecht: Reidel),109 Tsuge, K., Fukui, Y., Tachihara, K., et al. 2021, PASJ,73, S35 Tsuge, K., Tachihara, K., Fukui, Y., et al. 2021, PASJ,73, 417 Vollmer, B., Braine, J., Mazzilli-Ciraulo, B., et al. 2021, A&A,647, A138 Vollmer, B., Braine, J., & Soida, M. 2012, A&A,547, A39 Wang, Q. D., Owen, F., & Ledlow, M. 2004, ApJ,611, 821 Weisskopf, M. C., Tananbaum, H. D., Van Speybroeck, L. P., et al. 2000, Proc. SPIE,4012, 2 Whitmore, B. C., Brogan, C., Chandar, R., et al. 2014, ApJ,795, 156 Whitmore, B. C., Chandar, R., Schweizer, F., et al. 2010, AJ,140, 75 Whitmore, B. C., & Schweizer, F. 1995, AJ,109, 960 Whitmore, B. C., Zhang, Q., Leitherer, C., et al. 1999, AJ,118, 1551 Wilson, C. D., Scoville, N., Madden, S. C., et al. 2000, ApJ,542, 120 Xu, C. K., Lu, N., Condon, J. J., Dopita, M., & Tuffs, R. J. 2003, ApJ,595, 665 Yagi, M., Komiyama, Y., Yoshida, M., et al. 2007, ApJ,660, 1209 Yeager, T. R., & Struck, C. 2019, MNRAS,486, 2660 Yeager, T. R., & Struck, C. 2020a, MNRAS,492, 4892 Yeager, T., & Struck, C. 2020b, ApJ,905, 118 Zhu, M., Gao, Y., Seaquist, E. R., et al. 2007, AJ,134, 118 Zhu, M., Seaquist, E. R., & Kuno, N. 2003, ApJ,588, 243 23 The Astrophysical Journal, 931:121 (23pp), 2022 June 1 Appleton et al.