The Lyman Alpha Reference Sample: III. Properties of the Neutral ISM from GBT and VLA Observations
Stephen A. Pardy, John M. Cannon, Göran Östlin, Matthew Hayes, Thøger Rivera-Thorsen, Andreas Sandberg, Angela Adamo, Emily Freeland, E. Christian Herenz, Lucia Guaita, Daniel Kunth, Peter Laursen, J. M. Mas-Hesse, Jens Melinder, Ivana Orlitová, Héctor Otí-Floranes, Johannes Puschnig, Daniel Schaerer, Anne Verhamme
TThe Lyman Alpha Reference Sample: III. Properties of theNeutral ISM from GBT and VLA Observations
Stephen A. Pardy
Department of Astronomy, University of Wisconsin, 475 North Charter Street, Madison,WI 53706, USADepartment of Physics & Astronomy, Macalester College, 1600 Grand Avenue, Saint Paul,MN 55105, USA [email protected]
John M. Cannon
Department of Physics & Astronomy, Macalester College, 1600 Grand Avenue, Saint Paul,MN 55105 [email protected]
G¨oran ¨Ostlin
Department of Astronomy, Oskar Klein Centre, Stockholm University, AlbaNova UniversityCentre, SE-106 91 Stockholm, Sweden [email protected]
Matthew Hayes
Department of Astronomy, Oskar Klein Centre, Stockholm University, AlbaNova UniversityCentre, SE-106 91 Stockholm, Sweden [email protected]
Thøger Rivera-Thorsen
Department of Astronomy, Oskar Klein Centre, Stockholm University, AlbaNova UniversityCentre, SE-106 91 Stockholm, Sweden [email protected]
Andreas Sandberg
Department of Astronomy, Oskar Klein Centre, Stockholm University, AlbaNova UniversityCentre, SE-106 91 Stockholm, Sweden a r X i v : . [ a s t r o - ph . GA ] S e p [email protected] Angela Adamo
Max Planck Institute for Astronomy, K¨onigstuhl 17, D-69117 Heidelberg, Germany [email protected]
Emily Freeland
Department of Astronomy, Oskar Klein Centre, Stockholm University, AlbaNova UniversityCentre, SE-106 91 Stockholm, Sweden [email protected]
E. Christian Herenz
Leibniz-Institut f¨ur Astrophysik (AIP), An der Sternwarte 16, D-14482 Potsdam, Germany. [email protected]
Lucia Guaita
Department of Astronomy, Oskar Klein Centre, Stockholm University, AlbaNova UniversityCentre, SE-106 91 Stockholm, Sweden [email protected]
Daniel Kunth
Institut d’Astrophysique de Paris, 98bis, bd Arago, 75014 Paris, France [email protected]
Peter Laursen
Dark Cosmology Centre, Niels Bohr Institute, University of Copenhagen, Juliane MariesVej 30, 2100 Copenhagen, Denmark [email protected]
J. M. Mas-Hesse
Centro de Astrobiolog´ıa (CSIC–INTA), Madrid, Spain [email protected]
Department of Astronomy, Oskar Klein Centre, Stockholm University, AlbaNova UniversityCentre, SE-106 91 Stockholm, Sweden [email protected]
Ivana Orlitov´a
Geneva Observatory, University of Geneva, 51 Chemin des Maillettes, CH-1290 Versoix,SwitzerlandAstronomical Institute, Academy of Sciences of the Czech Republic, Bocni II, CZ-14131Prague, Czech Republic [email protected]
H´ector Ot´ı-Floranes
Instituto de Astronom´ıa, Universidad Nacional Aut´onoma de M´exico, Apdo. Postal 106,Ensenada B. C. 22800 Mexico [email protected]
Johannes Puschnig
Department of Astronomy, Oskar Klein Centre, Stockholm University, AlbaNova UniversityCentre, SE-106 91 Stockholm, Sweden [email protected]
Daniel Schaerer
Geneva Observatory, University of Geneva, 51, Ch. des Maillettes, CH-1290 Versoix,SwitzerlandUniversit´e de Toulouse; UPS-OMP; IRAP; Toulouse, France [email protected]
Anne Verhamme
Geneva Observatory, University of Geneva, 51, Ch. des Maillettes, CH-1290 Versoix,Switzerland [email protected]
ABSTRACT
We present new H I imaging and spectroscopy of the 14 UV-selected star-forming galaxies in the Lyman Alpha Reference Sample (LARS), aimed for adetailed study of the processes governing the production, propagation, and escapeof Ly α photons. New H I spectroscopy, obtained with the 100m Green BankTelescope (GBT), robustly detects the H I spectral line in 11 of the 14 observedLARS galaxies (although the profiles of two of the galaxies are likely confused byother sources within the GBT beam); the three highest redshift galaxies are notdetected at our current sensitivity limits. The GBT profiles are used to derivefundamental H I line properties of the LARS galaxies. We also present new pilotH I spectral line imaging of 5 of the LARS galaxies obtained with the Karl G.Jansky Very Large Array (VLA). This imaging localizes the H I gas and providesa measurement of the total H I mass in each galaxy. In one system, LARS 03(UGC 8335 or Arp 238), VLA observations reveal an enormous tidal structurethat extends over 160 kpc from the main interacting systems and that contains > M (cid:12) of H I . We compare various H I properties with global Ly α quantitiesderived from HST measurements. The measurements of the Ly α escape fractionare coupled with the new direct measurements of H I mass and significantlydisturbed H I velocities. Our robustly detected sample reveals that both totalH I mass and linewidth are tentatively correlated with key Ly α tracers. Further,on global scales, these data support a complex coupling between Ly α propagationand the H I properties of the surrounding medium. Subject headings: galaxies: ISM — galaxies: starburst — galaxies: kinematicsand dynamics — radio lines: galaxies
1. Introduction
The Lyman-alpha emission line (Ly α ) at 1216 ˚A fulfills several extremely important rolesin observations of the high-redshift ( z ) Universe. Recombination nebulae re-process ∼ / α is a resonance line and is readily scattered by neutral hydrogen atoms. This scat-tering may occur in the interstellar medium (ISM) of the galaxies themselves, or in thecircumgalactic medium (CGM) that immediately surrounds them. Scattering in the CGMcan lead to extended Ly α structures such as those found in high-z Ly α emitters (LAE) bySteidel et al. (2011) and Zheng et al. (2011) among others. Inside the ISM, the radiativetransport of Ly α depends strongly on dust extinction (Atek et al. 2014). This dependence,however, shows large scatter due to the complicated resonant scattering of Ly α , in which thevisibility of the line is influenced by a large number of factors, including dust content (Char-lot & Fall 1993; Verhamme et al. 2008; Atek et al. 2009; Hayes et al. 2010), dust geometry(Scarlata et al. 2009), neutral gas content and kinematics (Kunth et al. 1998; Mas-Hesse et al.2003; Cannon et al. 2004), and gas geometry (Neufeld 1991; Giavalisco et al. 1996; Hansen& Oh 2006; Laursen et al. 2013; Duval et al. 2014). Taken together, these myriad factorsindicate that the true total escape fraction of Ly α photons ( f Ly α esc ; defined as the ratio ofobserved to intrinsic Ly α luminosity; Hayes et al. 2005, Hayes et al. 2013), is still poorlyunderstood. Thus, photometric measurements of Ly α will reflect the underlying propertiesof galaxies only in the very broadest statistical sense.The only way to remedy this complicated situation is to study the Ly α emission linein a sample of star-forming galaxies on a spatially resolved basis. From the UV perspective,obtaining Ly α images in the nearby Universe is expensive and requires space-based platforms.To this end, our team has acquired data from the Hubble Space Telescope (HST) to produceLy α images of a statistically selected sample of star-forming galaxies in the low- z Universe(see ¨Ostlin et al. 2014); details about the image processing methods can be found in Hayeset al. (2009). This program, called the “Lyman Alpha Reference Sample” (“LARS”), includesHST Cycle 18 imaging and Cycle 19 spectroscopic observations of 14 galaxies. These HSTdatasets form the backbone of a comprehensive multi-wavelength observational campaignthat will allow great strides forward in our understanding of Ly α radiative transport. Sincethe LARS sample is selected on UV luminosity and H α equivalent width, and not on Ly α characteristics, the composite sample was designed to overlap with both high- z and low- z star-forming galaxies. LARS will form a critical local interpretive benchmark for studies ofhigh- z galaxies for decades to come, and is well placed to take advantage of the remainingHST lifetime and motivate future observations with JWST.The LARS program (including sample selection) is described in detail in ¨Ostlin et al.(2014) and Hayes et al. (2014), hereafter referred to as Papers I and II respectively. Pre-liminary results derived from the Ly α images of the LARS galaxies are presented in Hayes 6 –et al. (2013). That work showed that in many LARS galaxies, Ly α emission is prominenton physical scales that exceed those of both the massive stellar populations and the starformation regions that give rise to the Ly α photons; this trend is very prominent in thepanels of Figures 1, 2, and 3, which show optical and UV images of the LARS galaxies.This may point towards Ly α photons being resonantly scattered to large radii in most of theLARS galaxies, although other possibilities have been proposed, such as ionization causedby hot plasma (Ot´ı-Floranes et al. 2012). It is likely that multiple of the aforementionedphysical properties (dust content and geometry, neutral gas content, kinematics, geometry)are facilitating this efficient resonant scattering.In order to decouple the effects of dust and neutral gas kinematics and densities, theLy α images must be compared with spatially resolved information about the neutral ISM.While UV and optical absorption line spectroscopy offers one avenue to address this issue,such data are limited to those regions in the foreground of only the brightest continuumsources (usually the emission line nebulae themselves). While our team is pursuing suchanalyses (see Rivera-Thorsen et al., in preparation, and Ly α profile modeling from Orlitovaet al., in preparation), these avenues should be complemented by direct probes of the neutralinterstellar medium in all regions of the LARS galaxies. To this end, in this manuscript webegin a detailed exploration of the neutral hydrogen in the LARS galaxies by presenting newsingle-dish HI spectroscopy and spatially resolved interferometric H I imaging. We use thesedata to derive global properties and to study the properties of the H I gas on bulk scales.Previous spatially resolved H I observations of galaxies in which Ly α radiative trans-port has been studied in detail are few. While the sample of systems with spectroscopicinformation about Ly α is robust, there are comparatively few systems that have direct Ly α imaging (see the discussion in Hayes et al. 2005, Hayes et al. 2009, ¨Ostlin et al. 2009, andreferences therein). For those systems that do, there has, until the present work, been nosystematic investigation of the H I on a spatially resolved basis; only a few selected systemshave detailed H I studies in the literature.Cannon et al. (2004) presented H I observations of two of the most well-studied andnearby Ly α -emitting galaxies in the local Universe, ESO 338 − IG 004 (Tol 1924 − − IG 004 and IRAS 08339+6517. The close proximity of the com-panions suggests that the interactions were recent. Since both primary systems are Ly α emit-ters, these data support two possible interpretations of Ly α escape from starburst galaxies:either a) the bulk ISM kinematics provides the means of escape for Ly α photons by shiftingH I atoms out of resonance, or b) the H I is sufficiently clumpy on scales below our reso- 7 –lution to allow for efficient Ly α escape. These interpretations are further supported by H I (Bergvall & Jorsater 1988) and Ly α observations (Kunth et al. 1998) of ESO400-G43.In this work, we begin to extend this type of analysis to the entire LARS sample. Basicproperties of the LARS galaxies are summarized in Table 1. Figures 1, 2, and 3 show opticaland UV images of the 14 LARS galaxies, with the left and right columns displaying thecolor HST images and the SDSS color images respectively. The right column includes theimmediate surroundings, while the left column has smaller fields of view, and are scaled toshow detail in the galaxies.We organize this manuscript as follows. In § § I profiles of the 14 LARS galaxies, while § I images of 5 of the 14 LARS systems. We discuss eachsystem individually in §
5, and discuss various correlations between galaxy global propertiesin §
6. We draw our conclusions in §
7. Throughout this paper we adopt the value of H =70 ± − Mpc − from the averaged WMAP 7 year data as presented in Komatsu et al.(2011).
2. Observations, Data Reduction and Analysis2.1. GBT Data
We obtained data from the National Radio Astronomy Observatory 100m Robert C.Byrd Green Bank Telescope (GBT ) under program GBT/11A-057 (Legacy ID QO13; P.I.¨Ostlin). Data were acquired in four observing sessions in March and April of 2011. Positionswitching observations were performed for all sources; 32,768 channels over the 12.5 MHz oftotal bandwidth produce a spectral resolution of 381.469 Hz (0.08 km s − ch − ).All reductions were performed in the IDL environment , using the GBTIDL packagedesigned at NRAO. We first imported the raw data for each galaxy into GBTIDL usingstandard averaging techniques, with zenith opacity coming from atmospheric data fromthe NRAO CLEO weather system. The data was moderately contaminated by terrestrialradio frequency interference (RFI); bad data were identified by examining individual spectra,blanked, and interpolated. The National Radio Astronomy Observatory is a facility of the National Science Foundation operatedunder cooperative agreement by Associated Universities, Inc. Exelis Visual Information Solutions, Boulder, Colorado ∼ − . Flux and widthmeasurements were taken on this smoothed profile. Two regions were selected by hand foreach galaxy: the first around the profile, and the second in an area free from RFI (used asan RMS noise region).The redshifts of the LARS galaxies (see table 1) place some of them in proximity to abandpass filter in the 1.2-1.3 GHz region. This bandpass filter is designed to shield againsta local radar signal, but contributes higher than usual system temperature (T sys ) values(and therefore system noise) to calibrated data in this frequency range. Further, the T sys values fluctuated markedly from one spectral scan to the next. The affected galaxies were allnon detections (LARS 12, LARS 13, and LARS 14) and have markedly higher T sys values (inparticular LARS 13) than would otherwise be expected. It should also be noted that thesethree galaxies are at the largest redshifts in our sample.Systematically measuring the flux and width of irregular, low SNR, H I spectra canbe challenging. To do so, we have modified previously well known measurement techniquesand find that they are robust over the range of SNR and profile shapes in our data. Ourmeasurement routines take as inputs the smoothed spectrum and the region suspected ofcontaining the galaxy. In our initial tests of archival data for NGC 5291 (available in thedistribution of GBTIDL), we found that hand selecting regions of the profile could addadditional bias, causing the width and flux measurements to vary by 2-5%. Instead, weopted to use χ minimization techniques to fit Gaussian components (a single componentfit for galaxies with a single peak and a multi component fit for double-horned profiles) tothe smoothed spectrum. These components were used to guide later analysis and ensuredrepeatability in our results even with low SNR. These Gaussian fits serve only to guideanalysis, however, and the flux and width measurements follow the methodology describedin Springob et al. (2005). This method fits lines to the profile in the region over which it variesthe most (here between 15% and 85% of the peak value) and measures width properties fromthese fits. Springob et al. (2005) found that this reduced the dependence on noise on theirmeasurements versus direct measurements. We modified this method to respond better tolow SNR values and to handle single peaked data more accurately. We consider this methodto be the “best representation” of the width measurements for these spectra, although it isnot the only method (e.g., the busy function; Westmeier et al. 2014). We recorded two SNRs 9 –based on the peak-to-rms ratio and on the integrated flux to integrated rms ratio.The flux accuracy of GBT data using standard reduction techniques is quoted as 10%in the GBTIDL calibration guide. The primary error terms are the RMS fluctuations of thespectra in the region around the target and the subjectivity of selecting the baseline regionsand RMS/profile regions. In testing it was seen that increasing the order of the baselinegenerally led to increasing error. We thus adopt a 10% flux error for “good” profiles and aconservative 20% for “irregular” profiles.Special attention was paid to the error terms presented in Springob et al. (2005) asmodified from Schneider et al. (1990). The error term given there adds an error associatedwith the quality of the baseline fit. Because we have assumed a separate baseline error, weremove this term to get a more or less standard error term: (cid:15) statS = rms √ W ∆ V (1)where (cid:15) statS is the noise in units of Jy km s − , and W is the area of integration for the profile.The width at 50% of the maximum value (W ) results reported are highly dependent onthe accuracy of the method described in Springob et al. (2005) which, as mentioned in theabove discussion, breaks down at lower SNRs.The error for the width and systemic velocity measurements is based on the equationsfound in Schneider et al. (1990). σV = 1 . W − W )(SNR) − (2)This equation is doubled to give the error of the width measurements W and W , whichrefer to the width at 20% and 50% of the maximum, respectively. We obtained D-configuration Very Large Array H I spectral line imaging of five LARSgalaxies under program VLA/13A-181 (Legacy ID AC 1123; P.I. Cannon). These data wereobtained with a standard WIDAR correlator configuration that provides 16 MHz of totalbandwidth. 1,024 channels deliver a native spectral resolution of 3.3 km s − . The data wereacquired in March and April of 2013, during seven observing sessions; details are summarizedin Table 3. The National Radio Astronomy Observatory is a facility of the National Science Foundation operatedunder cooperative agreement by Associated Universities, Inc.
10 –Data reductions followed standard prescriptions in CASA and in AIPS . The widebandwidth and frequency range of these observations resulted in a significant amount ofradio frequency interference that was removed from the data. Bandpass and flux calibrationswere derived using either 3C147 or 3C286 (see Table 3). The gains and phases were thencalibrated using observations of the phase calibrators, which were typically separated by ∼ − . A Gaussian UVTAPER was applied to producecubes with beam sizes of ∼ (cid:48) ; the resulting cubes were then convolved to circular beamsizes. The beam sizes and rms noises of the final datacubes are summarized in Table 4.Moment maps were derived using standard blanking procedures. First, each datacubewas spatially smoothed to a circular beam size that is larger than the original. The resultingsmoothed cube was then blanked at the 2.5 σ level, where σ is the rms noise in line-freechannels of the cube. This blanked cube is then inspected by hand, and only features that arespatially coincident across two or more consecutive channels are deemed to be real emissionand kept in the final blanked cube. This cube is then used as a blanking mask against theoriginal, unconvolved cube. Moment maps are then derived using standard procedures. Weassume a 10% uncertainty on the overall calibration and absolute flux scale of the VLAimages.It is important to emphasize that all of the LARS galaxies are at distances of more than100 Mpc (z opt ≥ I work (e.g., 5 (cid:48)(cid:48) = 2.4 kpc at a distance of 100 Mpc), we are sensitive only to the bulk H I properties of the neutral gas in these galaxies. As such, the VLA images presented here areused only to study the global morphology and large-scale H I kinematics. A detailed studyof the H I on finer physical scales will require deep observations in more extended arrayconfigurations. Common Astronomy Software Application (CASA) is developed by the NRAO. Astronomical Image Processing System (AIPS) is developed by the NRAO.
11 – Ly α Data
We use global Ly α data for all 14 LARS galaxies as presented in Paper I and Paper II.The HST data were obtained with the Wide Field Camera 3 (WFC3) and Advanced Camerafor Surveys (ACS) using H α and H β filters, and a combination of U, B, and I bands for FUVcontinuum imaging. We also acquired data with the Solar Blind Channel using a bandpassdesigned to isolate Ly α emission.We reduced imaging data using MULTIDRIZZLE in the standard HST pipelines andused the Lyman-alpha extraction Software (LaXs) (Hayes et al. 2009) to continuum subtractthe Ly α UV data. We refer the reader to Paper I and II for details of this process. The finalglobal properties (see Paper I and table 5) are measured within apertures that correspondto twice the Petrosian radius ( η = 0.2; Petrosian 1976). This reduces the effect of noise onthe measurements at large radii.
3. GBT HI Global Profiles
Presented in Figure 4 are the fully smoothed and reduced GBT H I profiles of the LARSgalaxies. Plotted over these are the Gaussian fits in gray, solid lines in red showing the sidesof the profile used in the peak W calculations, and the peak W line itself shown as ahorizontal blue dotted line. A discussion of each galaxy is included below.The observed properties from our GBT observations of each galaxy can be found inTable 2 along with properties derived from these using methods described below. Thedistances of each galaxy are calculated from the standard LARS luminosity distances. TheH I mass in units of M (cid:12) is calculated via M HI = 2 . × D S HI where D is the distancein Mpc and S HI is the H I flux integral in units of Jy km s − . We will discuss the mass valuefor each galaxy in section 5. Overall we find that our detections and mass estimates show astrong dependence on distance.It is important to emphasize that the 8 (cid:48) primary beam of the GBT at 21 cm is sensitiveto all H I -bearing objects in the observed frequency range. Even for the closest LARS galaxy(LARS 01; D=120 ± (cid:48) beam subtends a projected circular area that is nearly280 kpc in diameter; the potential for contamination in the GBT beam is significant. Wethus examined SDSS images of the area surrounding each LARS galaxy for nearby objects.Objects with spectroscopic data available from SDSS that might have contributed to the Fruchter, A. and Sosey, M. et al. 2009, ”The MultiDrizzle Handbook”, version 3.0, (Baltimore, STScI)
12 –GBT flux are noted below. These objects have velocities within 300 km s − of the targetLARS galaxy and are within the 8 (cid:48) primary beamsize of the GBT. Optical images of thefields that contain possible confusing sources (LARS 06 and LARS 11) are shown in Figures5 and 6; these two sources are discussed in more detail in §
5. We used the SDSS g bandto convert to luminosity L g , which we compared with a log(M HI /L B ) value of − M (cid:12) of H I ; this increases to 10 for our furthest twosystems. This information is also presented in Table 2.The profiles in Figure 4 show a variety of H I structure. Some galaxies are simple singlepeak profiles (e.g, LARS 04, LARS 09), while others show complex and asymmetric profiles(e.g., LARS 01, LARS 03). Still other systems have H I spectra that are almost certainlyconfused; the classic double-horn profile of LARS 06 is likely due to the contribution ofUGC 10028, a large spiral galaxy located only 1 (cid:48) to the southeast of LARS 06 (see Figure 5).Various observed properties of each GBT H I profile, and the quantities derived fromthese, are tabulated in Table 2: column 1 identifies the galaxy number; column 2 is thesystemic velocity in km s − ; column 3 is the linewidth at 50% in km s − ; column 4 is thelinewidth at 20% in km s − ; column 5 is the velocity offset between the optical velocity andthe H I central velocity; column 6 is the the single-dish integrated flux in Jy km s − ; column 7is the mass from the single-dish measurements in units of 10 M (cid:12) ; column 8 is the SNR fromthe integrated flux and the peak; column 9 is the rms noise value of the GBT in units of mJy;column 10 gives the baseline polynomial fit used in the calibration of the GBT spectrum;column 11 is the classification of the profile (‘S’ for single-horned, ‘D’ for double-horned, ‘N’for no detection, ‘I’ for highly irregular shapes, or the presence of RFI near the signal region,and ‘C’ for confusion due to known nearby galaxies.)
4. VLA H I Images
The H I moment zero (representing H I mass surface density) and moment one (rep-resenting intensity-weighted velocity) images of the 5 LARS galaxies observed in program13A-181 are presented in Figures 7, 8, 9, 10, and 11. In panel (a) of each figure, a DigitizedSky Survey image is overlaid with contours of H I mass surface density (individual contourlevels are provided in the caption to each figure). The beam size is shown by a blue circle,while a red square shows the location of the HST pointing for which we have UV imaging(see discussion in Paper I and II). In panel (b) of each figure, the H I moment one image 13 –(representing intensity weighted velocity) is presented in a rainbow color format; the velocityscale of each galaxy is shown by a colorbar.Based on the distances and optical angular sizes of the LARS galaxies (of order 1 (cid:48) orless; see Figures 1, 2, and 3), one would expect the H I to be distributed on similar angularscales. At the resolution of these data (beam sizes between 59 (cid:48)(cid:48) and 72 (cid:48)(cid:48) ), the LARS galaxieswould thus appear as unresolved sources in the H I data cubes. As we discuss below, three ofthe LARS galaxies have H I morphologies consistent with this interpretation. Surprisingly,two of the sources have H I distributions that largely exceed the beam size; we discuss thesesources in detail below.It is important to stress that even in cases where galaxies are unresolved by an H I beam, one can still extract bulk characteristics of the neutral ISM because of the spectrallyresolved nature of the data. Detailed kinematic analyses (e.g., rotation curve work or spa-tially resolved position-velocity diagrams) will be unavailable in such cases. However, withinthe spatial and spectral resolution elements of the data, one can derive bulk constraints onthe motions of the H I gas.We use the VLA H I images of the 5 LARS galaxies presented here to constrain twocritical properties. First, the H I data allow us to localize the H I associated with the LARSgalaxy. This in turn allows us to accurately measure the total H I mass of each system.These interferometric measurements often recover less of the H I flux, and therefore mass,than the GBT measurements. We attribute this to the low SNR of H I in these galaxies,the possible presence of low surface brightness H I features missed by the VLA in extendedregions or velocity space, and flux contamination from other galaxies, although we suspectthat this is only relevant in the cases marked as confused.It is interesting to note that at the distances of the LARS galaxies, interferometric ob-servations are perhaps better suited to determine the H I mass than single-dish observations(for example, the 60 (cid:48)(cid:48) characteristic resolution of the present data are about a factor of threebetter than the H I beam of Arecibo, the largest single-dish radio telescope in the world,and a factor of eight better than our GBT resolution), simply because of the potential forcontamination in the single-dish beam. Second, the present data are sensitive to large-scaledistributions of H I gas, some of which is expected to be tidal in origin based on the opticalmorphologies of the LARS galaxies alone (see, e.g., Figures 1, 2, and 3).In Table 4 we present VLA H I quantities for our five targeted galaxies derived in thesame method as described above for the GBT data. Column 1 identifies the galaxy number;column 2 is the systemic velocity in km s − ; column 3 is the linewidth at 50% in km s − ;column 4 is the linewidth at 20% in km s − ; column 5 is the velocity offset between the 14 –optical velocity and the H I central velocity; column 6 is the the interferometric integratedflux in Jy km s − ; column 7 is the mass in units of 10 M (cid:12) ; column 8 is the rms noise valuein units of mJy/Beam; column 9 is the median global column density of H I from a columndensity map convolved with the beam size given in 10 cm − .
5. Discussion of Individual Galaxies
We now present discussion of each LARS target based on our new H I data and on theHST imaging presented in Hayes et al. (2013), Paper I, and Paper II. LARS 01 is a strong Ly α emitter, with a global EW of 46 ˚A (Paper I) and a moderatestar formation rate (a few M (cid:12) yr − ). The single-dish H I observations (see Figure 4) revealan asymmetric double-horn profile with peak SNR of 6.5 and linewidth at 50% of 160 km s − .We recover a total H I flux of 0.7 Jy km s − which at our distance of 118 Mpc gives an H I mass of 2.3 × M (cid:12) . A search of the SDSS field revealed no other objects within a ± − velocity range.Assuming that the H I profile is unconfused, the asymmetric line shape and offset fromthe optical velocity is suggestive of bulk outflow of H I . Consistent with this interpretationis the observation in Paper II that LARS 01 shows extended H α and H β emission. Further,preliminary results from COS spectroscopy show that within the COS aperture (2.5 (cid:48)(cid:48) diame-ter), large column densities of neutral hydrogen are outflowing at around −
128 km s − withrespect to the H α velocity. Note that LARS 01 will be studied in detail within Paper I, whileRivera-Thorsen et al. (in preparation) will present the COS analysis of the whole sample.Interstellar metal lines are blueshifted with respect to the systemic velocity of the galaxy,and there is a distinct lack of static gas in the COS aperture. This is consistent with therecent results from Wofford et al. (2013), whose analysis of the interstellar OI, Si II and CIIlines in this galaxy also yields an outflow of the neutral gas at −
130 km s − (not includingthe geocoronal velocity offset applied by these authors). This large shift is also seen in ourmeasurement of the systemic velocity of H I in LARS 01.As the closest LARS galaxy, future interferometric H I observations of LARS 01 will beespecially interesting. 15 – This galaxy is the strongest Ly α emitter (in terms of escape fraction) in the samplewith a Ly α EW ∼
82 ˚A, yet has one of the lowest (H α or UV-based) star formation rates.The GBT spectrum is double peaked (see Figure 4) and has a linewidth ∼
140 km s − . Thetotal single-dish H I flux is 0.7 Jy km s − , which gives a mass of 2.7 × M (cid:12) at our deriveddistance of 127 Mpc. We find no sources of potential contamination in the SDSS field.The VLA images of LARS 02 (see Figure 7) reveal a compact distribution of H I gas; theH I is only slightly resolved by the 59 (cid:48)(cid:48) (36.3 kpc) beam. The H I column densities peak at ∼ × cm − (although the marginally resolved nature of the source strongly suggests thatthe H I distribution is more localized, with correspondingly higher mass surface densities).There is evidence for coherent rotation in this source; the isovelocity contours shown in panel(b) of Figure 7 are parallel from ∼−
15 – 15 km s − of the systemic velocity of 8960 km s − .A coarse estimate of the dynamical mass of the system can be made by assuming that thisrotational velocity is occurring over the angular diameter of the beam, and that the totalprojected rotation velocity is symmetric about a dynamical center. The resulting M dyn =3.8 × M (cid:12) (which will increase for any non-zero value of disk inclination, but will decreasedepending on the actual size of the rotating component of the disk) is consistent with theinterpretation in Paper II of LARS 02 being among the more dwarf-like systems in LARS.We stress that this dynamical mass estimate is meant to be representative only. LARS 03 is the nuclear region of the southeastern galaxy in the spectacular Arp 238interacting pair. Note that Figure 1 shows the large-scale interacting morphology veryclearly in the SDSS panel, while the HST Ly α imaging covers only the southern of the twonuclei, and only a relatively small portion of the total interacting system. Paper II finds thatLARS 03 is a weak Ly α emitter. Interestingly, the Ly α luminosity increases as a function ofdistance from the source, suggesting that Ly α photons are readily being scattered to largedistances even though the Ly α extension parameter ( ξ Lyα , defined as the ratio between theLy α and H α Petrosian radii) is unity.Based on the interacting morphology of the system, we expect both an irregular globalH I profile and extended H I structure. Both of these expectations are borne out by the data.The GBT profile of LARS 03 (see Figure 4) is double-peaked and asymmetric. The linewidthexceeds 300 km s − at the 50% level, and there is evidence for low surface brightness H I emission on even larger velocity scales (W = 380 ±
62 km s − , and the profile shows weak 16 –evidence for H I emission over a range as large as ∼
600 km s − ).The H I morphology and kinematics of LARS 03 clearly indicate a violent interactionbetween the two galaxies. The H I emission is extended many beam sizes beyond the opticalgalaxy. As shown in Hayes et al. (2013), Ly α emission is found in the southeast of the twointeracting galaxies (although those authors point out that the nature of Ly α emission onlarger spatial scales is not constrained by data presently in hand). Our VLA images revealthat the H I surface density maximum is coincident with the Northwest of the two interactinggalaxies (although this is within one beam width at the present spatial resolution). Thisregion is not included in the present HST observations. The total H I mass in the interactingsystem is ∼ × M (cid:12) , and the average H I column density as revealed by our VLA beamis 4.8 × cm − . An enormous tidal feature, which is contiguous in velocity space from thetwo interacting galaxies, extends ∼
160 kpc to the west-southwest and contains ∼ × M (cid:12) of HI gas. Because of the large distance separating this component from the main bodyof LARS 03 we decide to remove this mass from the total H I mass estimate of LARS 03itself. This extended tidal structure is reminiscent of that in the local Ly α -emitting starburstgalaxies Tol 1924 −
416 and IRAS 08339+6517 found by Cannon et al. (2004), in which thelarge-scale neutral gas kinematics are interpreted to be critical for Ly α radiative transport.Although we do not probe the same physical scales in this work, we can conclude from thislarge tidal feature that the neutral gas is strongly disturbed.Even with these interesting H I kinematics and the obvious interacting nature of thesystem, which contributes to its luminous infrared nature, LARS 03 remains a relatively weakemitter of Ly α photons, that would be undetected in most high-redshift surveys. While theever increasing Ly α emission to higher radii is tantalizing, more data is needed to draw firmconclusions. LARS 04 has a strong single peaked H I profile with a width at 50% of 150 km s − (seeFigure 4). We classify it as an unconfused source; although SDSS J130757.13+542310.6 (V opt = 9714 km s − ) is located 5.56 (cid:48) to the southwest, it is not detected in our VLA imaging (seebelow). LARS04 has an irregular morphology in the UV and is a net Lya absorber, justshowing a weak asymmetric emission on top of a large, damped absorption profile within theHST COS aperture (2.5 (cid:48)(cid:48) in diameter). Paper II has shown that weak diffuse Ly α emissionis present, but scattered to very large galactocentric radii.We recover a total single-dish H I flux of 1.6 Jy km s − which yields an H I mass of 17 –7.3 × M (cid:12) at the adopted distance of 140 Mpc. The VLA H I imaging (see Figure 9)reveals extended emission with a median H I column density of 4.1 × cm − [maximumangular extent roughly twice that of the 71 (cid:48)(cid:48) (47.5 kpc) beam]; interestingly, this extendedH I component is in the same directional sense as the optical morphology. Higher spatial res-olution observations of LARS 04 with the VLA are both technically feasible (sufficient sourcebrightness) and will be very useful in assessing the localized morphology and kinematics.There is coherent rotation of the source apparent at this low spatial resolution. Specifi-cally, the isovelocity contours are mostly parallel over ∼
100 km s − (see Figure 9). Using thesame approach as for LARS 02 above, this implies a dynamical mass of M dyn = 1.4 × M (cid:12) (again, with no inclination correction and with the assumption of the H I rotating over onebeam diameter). However, based on the highly irregular optical morphology (which Figure 1shows to be suggestive of an interaction), it is likely that the localized neutral gas kinematicsare more complicated than simple disk rotation; the coarse beam size is likely smoothing outthe kinematic details of this system. LARS 05 is an edge-on galaxy (see Figure 1), with net Ly α absorption at very small radiiand emission at larger radii, and at the physical scales relevant in this work [see discussionin Paper II]. The GBT H I profile (see Figure 4) shows a very weak signal that is near thenoise level. While the SNR is low, the peaks in the spectrum may represent the top of thehorns of a typical double-horn profile that is expected for a highly inclined disk galaxy. Wederive an H I flux integral of 0.55 Jy km s − , which, at a distance of 140 Mpc, correspondsto an H I mass of 2.4 × M (cid:12) . This galaxy has no known companions or contaminants inthe GBT field of view. LARS 06 is an irregular galaxy that has the weakest H α and UV-based star formationrates in the entire LARS sample; it is a Ly α absorber on all physical scales. The opticalmorphology suggests an interaction scenario, with the southern tail being UV luminous butH α dim compared to the main star forming knots in the northern portion of the system.Interestingly, this is the LARS system with the lowest measured nebular abundance [seediscussion in Paper II].The GBT H I profile of the LARS 06 field has the highest SNR of any of the LARS 18 –galaxies. This is almost certainly due to confusion within the beam. As Figure 5 shows,LARS 06 is separated from the disk galaxy UGC 10028 by only ∼ (cid:48) (43 kpc at the adopteddistance of 148 Mpc). Although the redshift-derived velocity values differ for these twogalaxies by ∼
200 km s − , the H I velocity of the total system is centered directly at thevelocity of UGC 10028. This single-dish H I spectrum is a classical double horn profileindicative of the rotation of a massive and inclined H I disk; the width of the profile is ∼
370 km s − at 50% of the maximum. The H I flux of 4.4 Jy km s − yields a total H I massof 2.3 × M (cid:12) ; this can be compared with the estimate of the stellar mass of LARS 06,2.1 × M (cid:12) , from Paper II. While gas to stellar mass ratios of this size are not unreasonable,a comparison of the optical morphologies of UGC 10028 and LARS 06 strongly suggests thatthe former system is contributing significantly to the flux in this field. From the SDSS g-band luminosity, we estimate the H I mass of UGC 10028 as 4.1 × M (cid:12) . This representsroughly 20% of the total mass seen in the system, but we caution that this might be an overor under estimation if the galaxy does not follow the same relation as other galaxies.There are three other galaxies in the vicinity of LARS 06 that could also be contribut-ing to the observed flux integral. As discussed in the caption of Figure 5, LARS 06 andUGC 10028 are also in close angular proximity to 2MASX J15455278+4415470, 2MASXJ15455157+4415310, and SDSS J154549.41+441539.2. Based on the optical velocity, 2MASXJ15455278+4415470 appears to be roughly 1000 km s − in the background of LARS 06 andUGC 10028. Only photometric redshifts are available for 2MASX J15455157+4415310 andSDSS J154549.41+441539.2, and each appears to also be at larger distances than LARS 06and UGC 10028; emission line spectroscopy will be required to determine the absolute ve-locities of these systems.Wide-bandwidth VLA observations of this complex field will be able to localize theneutral gas components of each system; this is an ideal target for subsequent high spatialresolution observations. LARS 06 is also the only detected galaxy that lies significantly off ofour observed gas fraction-stellar mass relationship (see discussion in § LARS 07 is a near edge-on disk system that shows one of the highest Ly α equivalentwidths of the LARS sample. The GBT spectrum shown in Figure 4 is measured as a singlepeak profile with a linewidth of ∼
92 km s − at 50% of maximum; a second peak is separatedfrom this main component by ∼
120 km s − . Due to the low SNR of this putative secondpeak, we measure the H I properties of the source using only the higher SNR component. 19 –With this assumption, LARS 07 has the narrowest linewidth of any galaxy in our sample.The H I flux is 0.47 Jy km s − , corresponding to an H I mass of 2.9 × M (cid:12) at the adopteddistance of 161 Mpc. LARS 08 is a metal-rich, low-inclination system (see Figure 2 and discussion in PaperII). The distribution of UV, Ly α , and H α emission is strongly asymmetric in the galaxy,being much more prominent on the western side of the disk than on the eastern side. Thesystem is a Ly α emitter on all spatial scales.The GBT spectrum of LARS 08 (see Figure 4) shows a broad, possibly double-peakedprofile with a linewidth of 310 km s − at 50% of the peak. The single-dish H I integratedflux is 3.4 Jy km s − , yielding a total H I mass of 2.2 × M (cid:12) at the adopted distance of160 Mpc. The galaxy appears to be isolated, with no known neighbors in the GBT beamwith similar velocities. The median column density of H I is 2.5 × cm − .As Figure 10 shows, the detected H I emission is highly localized to the 72 (cid:48)(cid:48) (56.9kpc) beam. The source appears to be undergoing bulk rotation; while the velocity field isformally unresolved, the nearly parallel isovelocity contours span ∼
80 km s − (note that thefirst moment map represents intensity-weighted velocity; for faint, unresolved sources onewould expect a compressed velocity scale in the first moment compared to a spectrum).Taken as projected rotation with no inclination correction (which is likely substantial, giventhe apparent optical inclination), the implied dynamical mass is ∼ × M (cid:12) . Higherspatial resolution H I observations of LARS 08 would be very interesting to pursue, since thissystem appears to be undergoing normal rotation and lacks obvious signatures of large-scalekinematic disturbances in our H I maps, even though COS spectroscopy shows significantoutflow activity (see details in Rivera-Thorsen et al., in preparation). LARS 09 is an extended system with two prominent arms that are rich in H α -emittingstar formation regions (note that a careful inspection of Figure 2 shows that there is aprominent foreground star at the southern end of the disk). The optical morphology isconsistent with the source being either a loose spiral or an interacting pair. The system is aglobal Ly α emitter at large scales (R >
10 kpc); the Ly α morphology is similar to that ofthe optical/UV, although more extended in all directions from the disk. 20 –The GBT spectrum (see Figure 4) is broad (linewidth of 270 km s − at 50% of the peak)and relatively low SNR. The total H I single-dish flux is 1.2 Jy km s − , which gives a mass of1.2 × M (cid:12) at a distance of 200 Mpc. The profile is not contaminated by any other knownsources within the GBT beam. The median H I column density is 1.2 × cm − The VLA images presented in Figure 11 show that the source is essentially unresolved bythe 59 (cid:48)(cid:48) (57.2 kpc) beam. The H I maximum is co-spatial with the optical body; there is weakevidence for extended gas that is only slightly larger than the beam size (see, for example,the extended material to the east-southeast of the optical body in Figure 11). However, wedo not interpret this as evidence of extended tidal structure due to the marginal extensioncompared to the beam size.The intensity weighted velocity field of LARS 09 covers a large range ( >
200 km s − )and is severely disturbed. There are some isovelocity contours that are roughly parallelin the northwest region of the system; however, at the position of the optical body thesecontours deviate significantly from regularity. The kinematic information in these imagessuggests that the interaction scenario for the optical morphology may be appropriate. Thepresence of diffuse Ly α emission in this source, combined with the irregular kinematics andmorphology, make this a prime target for higher resolution H I observations. LARS 10 is an irregular source that shows high Ly α equivalent width emission arisingfrom a diffuse component that extends beyond the optical body of the source [see discussionin Paper I]. The optical morphology is consistent with an advanced merger state. The GBTprofile is broad ( ∼
280 km s − at 50% of maximum), although of low SNR; the total fluxintegral of 0.29 Jy km s − corresponds to a total H I mass of ∼ × M (cid:12) at the adopteddistance of 250 Mpc. The weak H I flux integral will make high resolution interferometricobservations of LARS 10 challenging. LARS 11 is a dramatic edge-on galaxy that is in close angular proximity to a field spiralgalaxy (CGCG 046-044 NED01 at 14 h m s , +06 ◦ (cid:48) (cid:48)(cid:48) ). LARS 11 appears to be in theforeground of this object (LARS 11 is at z opt = 0.0843, while CGCG 046-044 NED01 has aphotometric redshift derived from SDSS of ∼ h m s +06 ◦ (cid:48) (cid:48)(cid:48) , and SDSS J140353.36+062504.8at 14 h m s +06 ◦ (cid:48) (cid:48)(cid:48) . Given this complex field, we classify the GBT profile of LARS 11as potentially confused. Further, RFI was present near the expected frequency of the H I spectral line from LARS 11; this required extensive flagging.The H I profile of LARS 11 shown in Figure 4 is broad (W = 260 km s − ) and isdistributed in multiple peaks. This may represent a broad rotation profile from LARS 11,or it may be indicative of multiple sources contributing to the H I flux within the beam.Assuming that the full H I flux integral of 0.75 Jy km s − is only associated with LARS 11,the implied neutral hydrogen mass is 2.3 × M (cid:12) at the adopted distance of 360 Mpc.VLA imaging of LARS 11 would be very useful for localizing the H I emission from thevarious sources in this complex field.
6. Global Characteristics of the LARS Galaxies
Due to resonant scattering with neutral hydrogen atoms, the radiative transport ofLy α photons may well have its strongest dependence on the characteristics of the H I thatsurrounds the star forming regions in which the photons are produced. With the presentH I spectroscopy and imaging, we now have a first order understanding of the neutral gascontents of many of the LARS galaxies. Comparing the qualities of the global H I reservoirswith the detailed characteristics of each system derived from HST imaging allows us to probewhat roles H I kinematics and density play in governing Ly α radiative transport.We focus on four global properties derived from the present H I observations of the11 lowest-redshift LARS galaxies (those for which the GBT spectra provide meaningfulmeasurements or limits); for the galaxies where it is available (LARS 02,03,04,08,09) weuse the total fluxes and masses VLA images preferentially over the GBT data. In generalthe VLA data recovers 30% less flux than the GBT data, which contributes to 6% shorterlinewidths. As mentioned in section 4, this is most likely due to a combination of decreasedsurface brightness sensitivity in the VLA data and decreased contamination from othersources.First, the H I linewidth (specifically, the width at 50% of the maximum W ) is adistance-independent variable that has been shown to correlate with absolute magnitudeand thus to serve as a proxy for mass in star-forming disk galaxies (Tully & Fisher 1977).For rotationally dominated galaxies, a larger H I linewidth can often be interpreted as anindicator of a more massive galaxy. To know with certainty, a detailed study of the dynamics 22 –and complete censuses of the baryonic components of individual galaxies is needed. Inour case, the galaxies with irregular morphology may have linewidths that are significantlyincreased due to kinematic disturbances and merging activity.The second global parameter we examine is the total H I mass of each LARS galaxy.This parameter is of course related to the H I linewidth, but not in a one to one sense.The LARS galaxies span a range of morphological types, including violent interactions(e.g., LARS 03), disk-dominated spirals (e.g., LARS 05 and LARS 11), and irregulars (e.g.,LARS 04, LARS 06). While for disk dominated systems the H I linewidth and H I massshould be closely correlated, for the other types of systems the H I line profile may not onlybe indicative of rotation; H I gas from multiple components may create irregular single-dishprofiles. Examples of this are clearly seen in the GBT spectra (Figure 4) and in the VLAimages (see, e.g., Figure 8).Third, we examine the offset in velocity between the centroid of the H I profile andthe known systemic velocity of each LARS galaxy as derived from optical spectroscopy.Assuming that such offsets are caused by large-scale kinematic deviations from regularlyrotating galaxy disks (e.g., tidal interaction, gas outflow or infall), this offset parameter canbe considered a rough proxy for the presence or absence of large-scale motions of neutral gas.As discussed in Cannon et al. (2004) and above for LARS 03, we find evidence for extendedtidal structure in some Ly α -emitting galaxies. However, we also find systems that are strongLy α emitters that do not show such extended neutral gas components (e.g., LARS 02; see § I mass divided by the stellarmass. Numerous works have shown that normal, star-forming disk galaxies populate a robusttrend of decreasing gas fraction with increasing stellar mass (see recent results in Huang et al.2012, Papastergis et al. 2012, Peeples et al. 2014, and various references therein). In Figure 12we show the gas fraction as a function of total stellar mass (M ∗ ) derived from 2-componentSED modeling to the HST data (see Paper II and Hayes et al. 2009 for details). Exceptfor LARS 06, whose H I flux integral is likely strongly contaminated by the nearby fieldspiral UGC 10028 (see discussion in § I properties of LARS 06 arealmost certainly contaminated by UGC 10028, and those of LARS 11 may be contaminatedby SDSS J140401.00+062901.7 and/or SDSS J140353.36+062504.8. If these sources are infact contaminated, then the H I linewidths and masses will be overestimated; the velocity 23 –offsets may or may not change.We examine the relationships of these four H I -based quantities with seven global prop-erties derived from HST data in Paper II: the Ly α /H α flux ratio, the Ly α luminosity, theH α /H β flux ratio, the Ly α escape fraction, the Ly α equivalent width, the Ly α extensionparameter (Ly α ξ ), and the SFR per unit area within the Petrosian radius ( η = 0.2; seePaper II). As discussed in detail in Hayes et al. (2014), there is a complex interrelation be-tween these (and other) properties in the process of Ly α radiative transport; the total Ly α luminosity of a given galaxy does not correlate in a statistically significant way with anyindividual quantity. Further, there is significant variation in these values as a function ofposition within an individual galaxy; some galaxies appear as net absorbers within the diskand as net emitters when the diffuse Ly α halo is included. Nevertheless, we seek statisticalcomparisons with many of the global quantities of both Ly α and H I . These comparisonsmost closely match detections of higher-redshift Ly α emitters and will constrain one of themany poorly understood aspects of Ly α propagation and detection in the distant universe.From a simplified interpretative standpoint, these seven global quantities can be relatedas follows. The H α /H β ratio indicates extinction; deviations above the intrinsic ratio of 2.86(Osterbrock 1989) indicate non-zero E(B − V) values. These E(B − V) values can in turn beused to derive the intrinsic Ly α luminosity of a system based on its observed (and extinctioncorrected) H α luminosity and also to estimate the Ly α escape fraction. The Ly α /H α ratiohas a (Case B) recombination value of 8.7. Lower values can indicate stronger suppressionof Ly α versus H α (e.g., by attenuation from dust, or by resonant scattering away from theproduction site); super-recombination values are seen in a few localized regions of the LARSgalaxies, as well as in some of the large scattered Ly α halos surrounding some of the systems(see Paper I and Paper II). The Ly α equivalent width is related to the recent star formationhistory of the galaxy or region; global values of ∼
100 ˚A indicate constant star formation over ∼
100 Myr timescales, while values exceeding 250 ˚A occur only in the very youngest burstepisodes (Schaerer et al. 2003 and Raiter et al. 2010.)We present scatter plot comparisons of the four global H I -derived properties and sevenglobal HST-derived properties in Figures 13, 14, 16, and 15. We restrict the Ly α propertiesto positive values, which has the effect of setting EW and luminosity measurements equalto zero. We mark this region on the plots with a hashed region. We use the Spearman ρ s correlation coefficient to quantify possible monotonic correlations between these properties;a perfect correlation or anti-correlation between two properties will have ρ s values of +1 or −
1, respectively. Each panel of Figures 13 through 16 shows the corresponding ρ s value. Toestimate the errors on these correlation coefficients we resample the properties 1000 timeswith a random realization of the associated errors and measure, recording the correlation 24 –coefficients each time. For each correlation we show the value from the sample of uncontam-inated detections as the primary result, and include the value for the entire sample includingupper limits and confused detections below.Overall the results do not show strong evidence of correlation between properties ( ρ s < − ρ s > +0.6). We present all correlations in Table 6, along with the dispersion anddiscuss the correlations for each H I property below. The H I line width is significantly anti-correlated with the Ly α extension parameter, ξ Ly α (see figure 14). This is the strongest evidence of correlation in the entire sample. TheH I line width is positively correlated with the H α /H β ratio (see further discussion below).Two factors are possibly at play here. Either larger H I quantities scatter Ly α photonsto extended radii, or interactions increase both the linewidth and preferentially block Ly α photons at short radii. Two anti-correlations, and a positive correlation appear comparing the H I mass of oursample of well detected galaxies (see Figure 13). The escape fraction ( ρ s = − α EW ( ρ s = − α /H β ratio shows strong evidence ofcorrelation ( ρ s = 0.75). It has been known for some time that more massive star forminggalaxies (hence those with larger H I masses and linewidths) are dustier, and that theBalmer decrement scales accordingly (e.g., Brinchmann et al. 2004; Lee et al. 2009; Garn &Best 2010). How this correlation may be connected to the underling Ly α propagation is notentirely clear. We see no conclusive evidence of correlations with any of the global Ly α propertiesusing either the whole sample or the positive detections. The strongest correlation is withthe H α /H β ratio ( ρ s = -0.53). 25 – We find no significant correlation between the velocity offsets and the various Ly α properties. This might be due to our velocity smoothing and the large extent over whichwe are probing the velocity information. The strongest correlation is with with the H α /H β ratio ( ρ s = 0.50).
7. Discussion and Conclusions
We have presented new GBT spectroscopy and VLA H I spectral line imaging of thegalaxies in the “Lyman Alpha Reference Sample” (“LARS”). LARS is a comprehensive,multi-wavelength study of 14 UV-selected star-forming galaxies that functions as a localbenchmark for studies of Ly α emission and absorption at higher redshifts. The centerpieceis a suite of HST images that allows us to study Ly α emission and related quantities on aspatially resolved basis. The preliminary results presented in Hayes et al. (2013) showed thatmost of the LARS galaxies have large, diffuse halos of Ly α emission that exceed the sizesof the stellar populations; this is strong evidence for the importance of resonant scatteringin the neutral hydrogen gas component. The HI structure and content are important piecesfor understanding the origin of the Ly-alpha halos and their different sizes. Our current HIangular resolution probes the large-scale global properties rather than those in the immediatevicinity of the Ly-alpha halo regions imaged with HST. Nevertheless, these provide importantinsights into the galaxy structure and environment. The LARS survey products are describedin Paper I of this series ( ¨Ostlin et al. 2014); the integrated properties of the LARS galaxiesas derived from the HST imaging are presented in Paper II of this series (Hayes et al. 2014).In this manuscript we present a first exploration of the H I properties of the LARSgalaxies. Using data from the GBT 100 m telescope, we have presented direct measurementsof the neutral gas contents of 11 of the 14 galaxies; we place limits on the three most distant(z > I peaks. Two sources (LARS 06 and LARS 11)are likely confused with nearby galaxies. We fit each of these profiles in order to derive globalline properties, including systemic velocity, linewidths, H I flux integrals, and H I masses.For a subset of five of the LARS galaxies, we also present new, low-resolution H I spectral line imaging obtained with the VLA. These H I images allow us to localize the H I associated with each galaxy. Three of the systems (LARS 02, LARS 08, and LARS 09) areunresolved at this angular resolution, while LARS 04 is resolved by a few beams. LARS 03 26 –is highly resolved by the present data; we have discovered an enormous tidal structure thatcontains more than 10 M (cid:12) of H I , and that extends more than 160 kpc from the maininteracting galaxies in the LARS 03 system. Based on the recovered flux integrals from theseVLA data, each of the five systems presented here would be amenable to higher spatialresolution observations. The VLA C and B configurations would offer a factor of roughly 4and 10 improvement in resolution respectively, but at the cost of longer integration times.In particular LARS 03 presents an interesting case where we have unequivocal evi-dence of kinematic disturbances on large scales, yet no accompanying boost relative to non-disturbed galaxies in Ly α escape fraction. A high resolution follow up with the VLA andfuture radio telescopes, especially coupled with larger maps of the Ly α flux, would disen-tangle the local versus global kinematic effects, and probe the Ly α escape on physical scaleslarger than the ∼
10 kpc already probed in the HST imaging.Using these new GBT and VLA data, as well as the results derived from HST imagingand presented in Paper II, we compare various global H I and Ly α properties. While thesample is small, we find a few intriguing correlations in our robustly detected galaxies: TheH I linewidths are strongly anti-correlated with ξ Ly α ( ρ s = − I masses show anti-correlations with escape fraction and Ly α EW. Additionally, both the H I linewidths and the H I masses are each correlated with the H α /H β ratio. These intriguingbut tentative (anti-)correlations are in general agreement with a significant dependence ofLy α propagation on the total mass and thus the H I mass of the source, and is in generalagreement with previous work by Laursen et al. (2009) which found that the escape fractiondecreases with increasing virial mass. These results also support the general trend thatLy α properties appear to correlate with total galaxy mass, as found in Paper II. Thatwork showed that Ly α emission (quantified by Ly α equivalent width, escape fraction, etc.)is systematically larger in lower mass galaxies. Given the variety of properties that scaleroughly with mass (e.g., metallicity, dust content, star formation rate, UV colors, etc.), alarger statistical sample of galaxies will be needed in order to better quantify these trends.The famous metal-poor blue compact dwarf galaxy I Zw18 reminds us that the trendof lower mass galaxies having larger Ly α luminosities or escape fractions is not always true.While the system has an extremely low nebular oxygen abundance (Skillman & Kennicutt1993), an appreciable star formation rate (Cannon et al. 2002), and a substantial UV lumi-nosity (Grimes et al. 2009), Kunth et al. (1998) showed that I Zw 18 is in fact a Ly α absorber.This process was explained by numerous scattering events in the neutral, and, most impor-tantly, static ISM even in the complete absence of dust (Atek et al. 2009). Future local Ly α studies should seek to expand the number of low-mass systems with analyses similar to theones presented here for LARS. 27 –Although it is tempting to view the increasing Ly α /H α with decreasing mass as a resultof less gas mass being available for scattering, this theory must be cast in the light of resultsthat suggest that gas fraction increases with increasing redshift (e.g., Tacconi et al. 2013),and that lower mass galaxies tend to be more gas-rich (Saintonge et al. 2011) and less dusty(e.g., Blanc et al. 2011). In Figure 12 we see that the LARS galaxies fit the general trendwith gas mass fraction.While Ly α halos seem to appear in systems with a range of H I linewidths, largeLy α /H α values only appear in galaxies with H I smaller linewidths. Large Ly α halos arealso correlated with less massive galaxies, and with increasing velocity offsets. We interpretthis to mean that low H I linewidths are necessary but not sufficient for Ly α escape, whilelarger line widths contribute to the overall effects of Ly α destruction by dust, extinction,and scattering.The bulk of the evidence presented by previous Ly α work shows the importance of H I properties, especially kinematic details, in determining the escape of photons (e.g. Mas-Hesse et al. 2003). This process is mostly governed by local effects, and it is unclear at thistime what global H I properties are required for the escape and propagation of Ly α photons.The asymmetric GBT profiles and evidence of Ly α emission in the presence of large columndensities of H I both point to interesting sub-resolution kinematic and density effects.A larger sample of galaxies with data like those presented here will allow us to placethese correlations on a more statistically robust footing. The HST eLARS program has beenapproved for 54 orbits and contains 28 more galaxies within our nominal sensitivity rangeof z ∼ I and Ly α detected galaxies until newadvances of radio telescopes, such as the Square Kilometer Array , push H I detections to20-100 kpc scales at z > REFERENCES
Atek, H., Kunth, D., Hayes, M., ¨Ostlin, G., & Mas-Hesse, J. M. 2008, A&A, 488, 491Atek, H., Kunth, D., Schaerer, D., et al. 2009, A&A, 506, L1Atek, H., Kunth, D., Schaerer, D., et al. 2014, A&A, 561, A89Bergvall, N., & Jorsater, S. 1988, Nature, 331, 589Blanc, G. A., Adams, J. J., Gebhardt, K., et al. 2011, ApJ, 736, 31Brinchmann, J., Charlot, S., White, S. D. M., et al. 2004, MNRAS, 351, 1151Cannon, J. M., Skillman, E. D., Garnett, D. R., & Dufour, R. J. 2002, ApJ, 565, 931Cannon, J. M., Skillman, E. D., Kunth, D., Leitherer, C., Mas-Hesse, M., ¨Ostlin, G., &Petrosian, A. 2004, ApJ, 608, 768Charlot, S., & Fall, S. M. 1993, ApJ, 415, 580Duval, F., Schaerer, D., ¨Ostlin, G., & Laursen, P. 2014, A&A, 562, A52Finkelstein, S. L., Rhoads, J. E., Malhotra, S., Grogin, N., & Wang, J. 2008, ApJ, 678, 655Garn, T., & Best, P. N. 2010, MNRAS, 409, 421Giavalisco, M., Koratkar, A., & Calzetti, D. 1996, ApJ, 466, 831Grimes, J. P., Heckman, T., Aloisi, A., et al. 2009, ApJS, 181, 272Hansen, M., & Oh, S. P. 2006, MNRAS, 367, 979Hartmann, L. W., Huchra, J. P., Geller, M. J., O’Brien, P., & Wilson, R. 1988, ApJ, 326,101Hayes, M., ¨Ostlin, G., Mas-Hesse, J. M., et al. 2005, A&A, 438, 71Hayes, M., ¨Ostlin, G., Mas-Hesse, J. M., & Kunth, D. 2009, AJ, 138, 911Hayes, M., ¨Ostlin, G., Schaerer, D., et al. 2010, Nature, 464, 562Hayes, M., ¨Ostlin, G., Schaerer, D., et al. 2013, ApJ, 765, L27Hayes, M., ¨Ostlin, G., Duval, F., et al. 2014, ApJ, 782, 6Haynes, M. P., & Giovanelli, R. 1984, AJ, 89, 758 30 –Huang, S., Haynes, M. P., Giovanelli, R., et al. 2012, AJ, 143, 133Jester, S., et al. 2005, AJ, 130, 873Komatsu, E., et al. 2011, ApJS, 192, 18Kunth, D., Mas-Hesse, J. M., Terlevich, E., et al. 1998, A&A, 334, 11Laursen, P., Sommer-Larsen, J., & Andersen, A. C. 2009, ApJ, 704, 1640Laursen, P., Duval, F., ¨Ostlin, G. 2013, ApJ, 766, 124Lee, J. C., Gil de Paz, A., Tremonti, C., et al. 2009, ApJ, 706, 599Mas-Hesse, J. M., Kunth, D., Tenorio-Tagle, G., et al. 2003, ApJ, 598, 858Neufeld, D. A. 1991, ApJ, 370, L85Osterbrock, D. E. 1989, Research supported by the University of California, John SimonGuggenheim Memorial Foundation, University of Minnesota, et al. Mill Valley, CA,University Science Books, 1989, 422 p.,¨Ostlin, G., Hayes, M., Kunth, D., et al. 2009, AJ, 138, 923¨Ostlin, G.,Hayes, M., et al. 2014, ApJ, in preparationOt´ı-Floranes, H., Mas-Hesse, J. M., Jim´enez-Bail´on, E., et al. 2012, A&A, 546, A65Papastergis, E., Cattaneo, A., Huang, S., Giovanelli, R., & Haynes, M. P. 2012, ApJ, 759,138Peeples, M. S., Werk, J. K., Tumlinson, J., et al. 2014, ApJ, 786, 54Petrosian, V. 1976, ApJ, 209, L1Raiter, A., Schaerer, D., & Fosbury, R. A. E. 2010, A&A, 523, A64Saintonge, A., Kauffmann, G., Kramer, C., et al. 2011, MNRAS, 415, 32Scarlata, C., Colbert, J., Teplitz, H. I., et al. 2009, ApJ, 704, L98Schaerer, D. 2003, A&A, 397, 527Schlafly, E. F., & Finkbeiner, D. P. 2011, ApJ, 737, 103Schneider, S. E., Thuan, T. X., Magri, C., & Wadiak, J. E. 1990, ApJS, 72, 245 31 –Skillman, E. D., & Kennicutt, R. C., Jr. 1993, ApJ, 411, 655Springob, C. M., Haynes, M. P., Giovanelli, R., & Kent, B. R. 2005, ApJS, 160, 149Stark, D. P., Ellis, R. S., Chiu, K., Ouchi, M., & Bunker, A. 2010, MNRAS, 408, 1628Steidel, C. C., Bogosavljevi´c, M., Shapley, A. E., et al. 2011, ApJ, 736, 160Tacconi, L. J., Neri, R., Genzel, R., et al. 2013, ApJ, 768, 74Tully, R. B., & Fisher, J. R. 1977, A&A, 54, 661Verhamme, A., Schaerer, D., Atek, H., & Tapken, C. 2008, A&A, 491, 89Westmeier, T., Jurek, R., Obreschkow, D., Koribalski, B. S., & Staveley-Smith, L. 2014,MNRAS, 438, 1176Wofford, A., Leitherer, C., & Salzer, J. 2013, ApJ, 765, 118Zheng, Z., Cen, R., Weinberg, D., Trac, H., & Miralda-Escud´e, J. 2011, ApJ, 739, 62
This preprint was prepared with the AAS L A TEX macros v5.2.
32 –Table 1. Basic Properties of the LARS Sample
Galaxy Alternate RA Dec V opta z opta Distance b Morphological Type c name (J2000) (J2000) (km s − ) (Mpc)LARS 01 Mrk 259 13:28:44.50 +43:55:50 8394 0.028 120 Dwarf IrregularLARS 02 SDSS J090704.88+532656.6 09:07:04.88 +53:26:56 8934 0.030 130 Dwarf IrregularLARS 03 Arp 238 13:15:35.60 +62:07:28 9204 0.031 130 MergerLARS 04 SDSS J130728.45+542652.3 13:07:28.45 +54:26:52 9743 0.033 140 IrregularLARS 05 Mrk 1486 13:59:50.91 +57:26:22 10133 0.034 150 Edge-on Dwarf SpiralLARS 06 KISSR 2019 15:45:44.52 +44:15:51 10223 0.034 150 Dwarf IrregularLARS 07 IRAS 1313+2938 13:16:03.91 +29:22:54 11332 0.038 170 Dwarf Edge-on SpiralLARS 08 SDSS J125013.50+073441.5 12:50:13.50 +07:34:41 11452 0.038 170 SpiralLARS 09 IRAS 0820+2816 08:23:54.96 +28:06:21 14150 0.047 210 Edge-on SpiralLARS 10 MRk 0061 13:01:41.52 +29:22:52 17208 0.057 260 MergerLARS 11 SDSS J140347.22+062812.1 14:03:47.22 +06:28:12 25302 0.084 380 Edge-on SpiralLARS 12 SBS 0934+547 09:38:13.49 +54:28:25 30608 0.102 470 Dwarf IrregularLARS 13 IRAS 0147+1254 01:50:28.39 +13:08:58 43979 0.147 700 IrregularLARS 14 SDSS J092600.40+442736.1 09:26:00.40 +44:27:36 54172 0.181 880 Dwarf a Derived from SDSS spectroscopy. b Values derived from luminosity distance. c Morphologies provided to guide the reader and are presented from optical imaging and Paper II. Irregulars have no obviousspiral structure in the optical disk and mergers have an obvious interacting companion. Galaxies marked as dwarves have stellarmasses lower than 10 M (cid:12) .
33 –Table 2. Observed and Derived GBT H I Properties
Galaxy V sysa W ,c W Velocity Offset S
HId M HI SNR RMS Baseline e Type f (km s − ) (km s − ) (km s − ) (km s − ) (Jy km s − ) (10 M (cid:12) ) (Sum, Peak) (mJy) Fit Order(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11)LARS 01 8339 ± ±
10 180 ± − ± ± ± ± ±
17 150 ±
26 3 ± ± ± ±
20 310 ±
39 380 ±
62 230 ±
20 1.6 ± ± ±
16 150 ±
31 260 ± − ±
16 1.6 ± ± ±
14 160 ±
28 170 ± − ± < < ±
12 370 ± ± ± ± ± ±
16 100 ±
32 170 ± − ±
16 0.47 ± ± ±
23 310 ±
47 470 ±
73 20 ±
23 3.4 ± ± ±
46 270 ±
92 490 ± − ±
46 1.2 ± ± ±
69 280 ±
140 380 ±
220 70 ±
69 0.29 ± ± ±
56 260 ±
11 290 ±
18 65 ± ± ± < <
150 N/A 3.4 3 NLARS 13 (43980) (264) (290) N/A < < < <
310 N/A 2.0 1 N a Values from non-detections are from the SDSS optical redshifts. b These values are derived from fits to the slopes on either side of the profile. See discussion in § c Assumed linewidths for non detections are the average linewidth of detected sources. d Upper limits are 1.5 σ above the local rms noise over the average W value (290 km s − ). e Order of polynomial that was fit to the GBT baseline. f The classification of the profile: ’S’ for single-horned, ’D’ for double-horned, ’C’ for confusion caused by other galaxies, ’I’ for an irregular profile, ’N’for no detection.
34 –Table 3. VLA Observations of the LARS GalaxiesGalaxy Observation Primary Phase Flux IntegrationDate Calibrator Calibrator Density a Time(UT) (Jy) (min.)(1) (2) (3) (4) (5) (6)LARS 02 2013 Apr 3-4 0542+498=3C147 J0834+5534 8.28 ± ± ± ± ± ± ± a Flux density of phase calibrator derived during calibration. 35 –Table 4. Observed and Derived VLA H I Properties
Galaxy V sys W W Velocity Offset S HI M HI RMS (cid:104) N HI (cid:105) c (km s − ) (km s − ) (km s − ) (km s − ) (Jy km s − ) (10 M (cid:12) ) (mJy/Beam b ) (10 cm − )(1) (2) (3) (4) (5) (6) (7) (8) (9)LARS 02 8960 ± ± ± ± ± ± ±
10 170 ±
20 300 ±
32 316 ±
32 2.2 ± ± d ± ± ± ± ± ± ± ± ± ± ± ± ± ±
18 330 ±
28 -79 ±
28 0.60 ± ± a These values are derived from fits to the slopes on either side of the profile. See discussion in § b The final circular beam sizes of the VLA data presented here are 59 (cid:48)(cid:48) , 62 (cid:48)(cid:48) , 71 (cid:48)(cid:48) , 72 (cid:48)(cid:48) , and 59 (cid:48)(cid:48) , for LARS 02, LARS 03, LARS 04,LARS 08, and LARS 09, respectively. c Average global value after convolving column density map to the resolution of our beam. d Mass of main component of galaxy not including tidal tail.
36 –Table 5. Global UV properties of the LARS Sample Galaxies aGalaxy L Ly α L Hα f Ly αesc SFR
FUVcorr.
Metallicity W Ly α ξ Ly α M ∗ (10 cgs.) (10 cgs.) (M (cid:12) yr − ) (12+log(O/H)) (˚ A ) (10 M (cid:12) )(1) (2) (3) (4) (5) (6) (7) (8) (9)LARS 01 0.85 ± ± ± ± ± ± . +1 . − . ± ± ± ± ± ± ± . +0 . − . ± ± ± ± ± ± ± . +0 . − . ± ± ± ± ± ± ± ± ± ± ± . +0 . − . ± ± ± ± ± ± ± ± ± ± ± . +0 . − . ± ± ± ± ± ± ± . +0 . − . ± ± ± ± ± ± ± > ± ± ± ± ± ± ± . +0 . − . ± ± ± ± ± ± ± . +0 . − . ± ± ± ± ± ± ± . +0 . − . ± ± ± ± ± ± ± . +0 . − . ± ± ± ± ± ± ± . +0 . − . ± a All values from Hayes et al. (2013), ¨Ostlin et al. (2014) and Hayes et al. (2014).
37 –Table 6. Correlations between H I and global Ly α properties aLy α W M HI Velocity Offset f gas
Property ρ s a ρ s a ρ s a ρ s a (1) (2) (3) (5) (4)Ly α /H α − − − α ) − α /H β − Ly αescp − − − α EW − − ξ Ly α − − − − − − − a Values are for all galaxies with a positive, unconfused detection. Standarddeviation values are given in square brackets.
38 –Fig. 1.— Optical and UV views of LARS 01 – LARS 05. In the left column are colorcomposite images of the LARS galaxies from HST imaging (red: optical continuum, green:UV continuum, blue: Ly α , see Paper II). The number in the upper left panel corresponds tothe LARS identification number (e.g., 01 is LARS 01) and each panel in a row is the sameLARS galaxy. In the right column are SDSS color images of the LARS target galaxies andthe surrounding areas. The fields of view (which are tailored in each frame to show detail) islarger in the right column (1.69 (cid:48) on a side). B band magnitudes, given in blue, were obtainedby converting SDSS magnitudes using equations provided by SDSS and found in Jester et al.(2005). These values are corrected for foreground extinction using values from Schlafly &Finkbeiner (2011). 39 –Fig. 2.— Same as Figure 1, for LARS 06 – LARS 10. 40 –Fig. 3.— Same as Figure 1, for LARS 11 – LARS 14. 41 –Fig. 4.— GBT H I spectra of LARS Galaxies. We show each smoothed spectrum in blackover a gray Gaussian best fit line. The profile is integrated over the area where this bestfit rises above the background. The blue dashed line shows the W line derisved from themethod in Springob et al. (2005) and utilizing the sides of the profile shown by the two redlines on the sides. Refer to § opt = 10246 km s − ); UGC 10028 (V opt = 10399km s − ); 2MASX J15455278+4415470 (V opt = 11984 km s − ); 2MASX J15455157+4415310(SDSS RF method photometric redshift = 0.100 ± ± ± (cid:48) GBTprimary beam; labels denote the target source as well as possible contaminating galaxies atsimilar velocities within the primary beam. 44 –Fig. 7.— H I and optical comparison of LARS 02. Panel (a) shows a Digitized Sky Survey im-age, overlaid with contours of H I surface density at levels of (0.65,1.3,2.6,5.2,10.4,20.8) × cm − . The beam size (59 (cid:48)(cid:48) ) is shown in the upper left, and the approximate location and sizeof the 14 kpc x 14 kpc HST UV imaging is shown by a red square (see images and discus-sion in ¨Ostlin et al. 2014 and Hayes et al. 2014). Panel (b) shows the H I intensity-weightedvelocity field. 45 –Fig. 8.— H I and optical comparison of LARS 03. Panel (a) shows a Digitized Sky Survey im-age, overlaid with contours of H I surface density at levels of (0.65,1.3,2.6,5.2,10.4,20.8) × cm − . The beam size (62 (cid:48)(cid:48) ) is shown in the upper left, and the approximate location and sizeof the 13 kpc x 13 kpc HST UV imaging is shown by a red square (see images and discus-sion in ¨Ostlin et al. 2014 and Hayes et al. 2014). Panel (b) shows the H I intensity-weightedvelocity field. 46 –Fig. 9.— H I and optical comparison of LARS 04. Panel (a) shows a Digitized Sky Survey im-age, overlaid with contours of H I surface density at levels of (0.65,1.3,2.6,5.2,10.4,20.8) × cm − . The beam size (71 (cid:48)(cid:48) ) is shown in the upper left, and the approximate location and sizeof the 18 kpc x 18 kpc HST UV imaging is shown by a red square (see images and discus-sion in ¨Ostlin et al. 2014 and Hayes et al. 2014). Panel (b) shows the H I intensity-weightedvelocity field. 47 –Fig. 10.— H I and optical comparison of LARS 08. Panel (a) shows a Digi-tized Sky Survey image, overlaid with contours of H I surface density at levels of(0.65,1.3,2.6,5.2,10.4,20.8) × cm − . The beam size (72 (cid:48)(cid:48) ) is shown in the upper left,and the approximate location and size of the 20 kpc x 20 kpc HST UV imaging is shown bya red square (see images and discussion in ¨Ostlin et al. 2014 and Hayes et al. 2014). Panel(b) shows the H I intensity-weighted velocity field. 48 –Fig. 11.— H I and optical comparison of LARS 09. Panel (a) shows a Digi-tized Sky Survey image, overlaid with contours of H I surface density at levels of(0.65,1.3,2.6,5.2,10.4,20.8) × cm − . The beam size (59 (cid:48)(cid:48) ) is shown in the upper left,and the approximate location and size of the 26 kpc x 26 kpc HST UV imaging is shown bya red square (see images and discussion in ¨Ostlin et al. 2014 and Hayes et al. 2014). Panel(b) shows the H I intensity-weighted velocity field. 49 –Fig. 12.— Gas fraction versus total stellar mass, where M HI is the mass of H I and M ∗ isderived from 2-component SED modeling to HST data (see Hayes et al. 2014). The outlier isLARS 06, whose H I mass measurement is almost certainly contaminated by the nearby fieldspiral UGC 10028 (see Figure 5). We also show the correlations derived by Papastergis et al.(2012) and Peeples et al. (2014). It should be noted that these lines also take into accountmolecular gas, whereas our galaxies do not have molecular gas values. 50 –Fig. 13.— Comparisons of global H I mass with the seven global properties derived fromHST imaging, including Ly α /H α , L L α , H α /H β , f Ly α esc , Ly α EW, Ly α ξ , and UV SFR/R Ly αP .When available, we use the VLA derived values for H I mass, which are generally lower thanthe GBT derived values. The Spearman ρ s correlation coefficient, which quantifies possiblemonotonic correlations between properties, is shown in each panel. We compare these resultswith data from another local Ly α emitter, IRAS 08339+6517. The symbols represent thedifferent galaxies and are shown in the legend in the final panel. The same symbols are usedthroughout the rest of the comparison plots. 51 –Fig. 14.— Comparisons of global H I line width at 50% of maximum (see column 3 of Table 2)with the seven global properties derived from HST imaging, including Ly α /H α , L L α , H α /H β , f Ly α esc , Ly α EW, Ly α ξ , and UV SFR/R Ly αP . The Spearman ρ s correlation coefficient is shownin each panel for the subsample of galaxies with unconfused detections and, in parenthesis,all fourteen galaxies. The symbols are the same as in Figure 13. 52 –Fig. 15.— Comparisons between the gas fraction (M H I /M ∗ ) and the seven global propertiesderived from HST imaging, including Ly α /H α , L L α , H α /H β , f Ly α esc , Ly α EW, Ly α ξ , and UVSFR/R Ly αP . The symbols and correlation coefficients are plotted the same as in Figure 13. 53 –Fig. 16.— Comparisons between the velocity offset between optical and H I derived systemicvelocities and the seven global properties derived from HST imaging, including Ly α /H α , L L α ,H α /H β , f Ly α esc , Ly α EW, Ly α ξ , and UV SFR/R Ly αP20