Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Spatially resolved metal gas clouds

Spatially resolved metal gas clouds Spatially Resolved Metal Gas Clouds 1 1? 1;2 3 1;3 C. P eroux , H. Rahmani , F. Arrigoni Battaia and R. Augustin Aix Marseille Universit e, CNRS, LAM (Laboratoire d'Astrophysique de Marseille) UMR 7326, 13388, Marseille, France. GEPI, Observatoire de Paris, PSL Universit e, CNRS, 5 Place Jules Janssen, 92190 Meudon, France. European Southern Observatory (ESO), Karl-Schwarzschild-Str.2, D-85748 Garching b. Munc  hen, Germany. Accepted 2018 May 18. Received 2018 May 17; in original form 2018 March 5 ABSTRACT We now have mounting evidences that the circumgalactic medium (CGM) of galaxies is polluted with metals processed through stars. The fate of these metals is however still an open question and several ndings indicate that they remain poorly mixed. A powerful tool to study the low-density gas of the CGM is o ered by absorption lines in quasar spectra, although the information retrieved is limited to 1D along the sightline. We report the serendipitous discovery of two close-by bright z =1.148 extended galaxies with a fortuitous intervening z =1.067 foreground absorber. MUSE gal abs IFU observations spatially probes kpc-scales in absorption in the plane of the sky over a total area spanning 30 kpc . We identify two [O ii] emitters at z down to 21 kpc with SFR2 M /yr. abs We measure small fractional variations (<30%) in the equivalent widths of Fe ii and Mg ii cold gas absorbers on coherence scales of 8kpc but stronger variation on larger scales (25kpc). We compute the corresponding cloud gas mass <2 10 M . Our results indicate a good eciency of the metal mixing on kpc-scales in the CGM of a typical z1 galaxy. This study show-cases new prospects for mapping the distribution and sizes of metal clouds observed in absorption against extended background sources with 3D spectroscopy. Key words: galaxies: ISM { quasars: absorption lines { intergalactic medium 1 INTRODUCTION also Bird et al. 2015). These authors predict a mismatch be- tween the smoothing scales of H i and high-ionisation metals Baryons from the cosmic web are known to accrete ef- which has yet to be witnessed with observations. These nd- ciently onto galaxies. This mechanism sustains violent ings question how traditional absorption lines studies recover episodes of star formation which power out ows extending the true gas properties. Based on hydrodynamical equilib- out to the surrounding circumgalactic medium (CGM), and rium arguments, McCourt et al. (2018) further suggest that even reaching the larger scales of the intergalactic medium the CGM of galaxies cools via "shattering", resulting in a or IGM (Aguirre et al. 2001; Oppenheimer & Dav e 2006). high covering fraction of pc-scale, photoionised cloudlets. The metals carried by the out ows will either rain back Important issues thus remain unsolved: On which scales on to galaxies or get mixed into the IGM. Indeed, obser- are metals mixed? How does it vary with environment (i.e. vations of the IGM indicate signi cant quantities of met- IGM vs. CGM)? Do low and high-ionisation ions have dif- als at all redshifts (Pettini 2003; Ryan-Weber et al. 2009; ferent coherence scales? Can we nd direct observational D'Odorico et al. 2013; Shull et al. 2014). However, the mix- evidence of shattering on pc-scales? Where are the high- ing could remain incomplete (e.g. Dedikov & Shchekinov metallicity, intergalactic gas clouds? If the H i absorption 2004). Speci cally, Schaye et al. (2007) show that ionised and high-ionisation metal absorption gas arise in distinct metal clouds are compact (typical scales of 100pc) and physical gas structures, the observational techniques em- are short-lived. Once in the IGM they expand until they ployed to infer metallicities and the total mass of the warm- reach pressure equilibrium with their environment but re- hot CGM gas would be challenged (Tumlinson et al. 2011; main poorly mixed on scales of 1kpc or smaller. More re- Werk et al. 2014). The pockets of metal-rich material have cently, Churchill et al. (2015) nd that in the CGM of a also profound implications on metal-cooling eciency and in simulated dwarf galaxy, low ionization gas arises cloud struc- turn galaxy formation. Our understanding of these phases tures of scales of order  3kpc. High ionization gas how- of the gas and their metallicities has so far been limited by ever lies in multiple extended structures spread over 100 the lack of observational constraints. kpc and due to complex velocity elds, highly separated A powerful tool to study this low-density gas is o ered structures give rise to absorption at similar velocities (see by absorption lines in quasar spectra. In these quasar ab- sorbers, the minimum gas density that can be detected is E-mail:celine.peroux@gmail.com set by the brightness of the background source and thus Downloaded from https://academic.oup.com/mnrasl/advance-article-abstract/doi/10.1093/mnrasl/sly090/5004870 by Ed 'DeepDyve' Gillespie user on 08 June 2018 2 C. P eroux et al. the detection eciency is independent of redshift. High- zQSO=3.33 z =1.15 BG Background Quasar Background Galaxies quality quasar absorption spectra have produced a wealth of information regarding the distribution of heavy ele- ments (e.g. Kulkarni et al. 2005; Quiret et al. 2016). How- ever, the brightest background sources (quasars and gamma- ray bursts) are point-sources so that the observer is lim- ited to the information gained along the line-of-sight. The metallicity we typically infer from absorption studies is then zFG=1.07 not determined by the abundances of heavy elements on Foreground Galaxies the size of the metal concentrations, but by the metallic- zabs=1.07 ity smoothed over the size of the HI absorber, which is absorbers in typically 100 kpc (Bechtold et al. 1994; Schaye et al. 2003; Background Galaxy spectra Schaye & Aguirre 2005). On smaller scales the distribu- tion of metals is essentially unknown. To remedy this, ob- Figure 1. Sketch of the system lay-out. A bright z =3.33 quasar qso servers have used close quasar pairs (Hennawi et al. 2006; was the original target of the observations. In these data, we have Martin et al. 2010; Rubin et al. 2015, 2018) as well as mul- serendipitously discovered an intervening z =1.067 metal ab- abs tiple images from gravitationally lensed background objects sorber in the spectrum of two close-by z =1.148 extended back- gal to probe the transverse small-scale coherence along lines-of- ground galaxies. In addition, two emitting galaxies are observed sight dozen kpc apart (Rauch et al. 2001; Ellison et al. 2004; in [O ii] at the redshift of the absorber. Lopez et al. 2007; Chen et al. 2014; Rubin et al. 2017). By using an extended galaxy as background source however, one can directly map the distribution and sizes and consisted of three exposures of 900 s each. The sub- of the metal absorbers on small scales. By using this set- exposures were rotated by 90 degrees to minimise residuals up, Cooke & O'Meara (2015) have estimated that indeed from uneven at- elding. The eld of view is 60 "  60 ", high-column density neutral gas can span continuous areas a 0.2 "/pixel scale and a spectral sampling of 1.25 A/pixel 8 10 10 {10 times larger than previously explored in quasar or covering 4750{9350A. gamma-ray burst sightlines. Bergeron & Boiss e (2017) have The data were reduced with version v1.6.4 of the ESO used intervening absorber lines redshifted on more extended MUSE pipeline (Weilbacher 2015) and additional external quasar emission lines to probe the spatial covering of the gas routines for sky subtraction as explained in the following. clouds. In a remarkable work, Lopez et al. (2018) reported Master bias, at eld images and arc lamp exposures were Mg ii absorption along a bright lensed arc probing scales of produced based on data taken closest in time to the sci- the order 2{4kpc. They nd that the strength of the absorber ence frames. The raw science and standard star cubes were decreases with radius from the emitting galaxy as expected then processed correcting the wavelength calibration to a from the quasar absorber population. However the physical heliocentric reference. We checked the wavelength solution properties derived from the observations of lensed systems using the known wavelengths of the night-sky OH lines and rely heavily on the lensed model to compute the magni ca- nd it to be accurate within 18 km/s. The individual expo- tion factors and survey di erent physical area in the image sures were registered using the central quasar to ensure ac- plane, thus probing inhomogeneous ux levels. curate relative astrometry. Finally, the individual exposures Here, we report the serendipitous discovery of two were combined into a single data cube. The removal of OH bright z =1.148 extended galaxies with a fortuitous inter- gal emission lines from the night sky was accomplished with an vening foreground absorber at z =1.067 along their sight- abs additional purpose-developed code tested in previous work lines. The manuscript is organised as follows: Section 2 pro- (P eroux et al. 2017). After selecting sky regions in the eld, vides the observational set-up and details of the system lay- we created Principle Component Analysis (PCA) compo- out. Section 3 presents the constraints on the physical con- nents from the spectra which were further applied to the sci- ditions of the metal gas clouds. Finally, in section 4, we ence datacube to remove sky line residuals (Husemann et al. review the impact of these ndings in the broader context 2016, trouble). The seeing of the nal combined data was of the metal mixing in the CGM of galaxies. Throughout measured from the quasar. The resulting point spread func- 1 1 this paper we adopt an H = 70 km s Mpc , = 0:3, 0 M tion (PSF) has a full width at half maximum of 0:74" at the and = 0:7 cosmology. At the redshift of the absorber wavelength of the Mg ii absorber (5780 A), corresponding (z =1.067), 1" corresponds to 8.2 kpc. abs to 6 kpc at z =1.067 or 0:68" at 7700A. abs In this MUSE cube, we report the serendipitous dis- covery of two nearby z =1.148 [O ii] emitters, coined BGa gal and BGb, with bright continua (R mag=21.9 and 23.3 re- 2 MUSE OBSERVATIONS OF A spectively). The spectra of these background galaxies show REMARKABLE SYSTEM evidences of a "down-the-barrel" out ow with strong absorp- MUSE observations of the eld of quasar SDSS J02500757 tions in Mg ii, Mg i and Fe ii typical of bright galaxies at z=1 were undertaken in service mode in natural seeing mode un- (Kornei et al. 2012; Martin et al. 2012). A more remarkable der programme 095.A-0615(A). These observations are part feature however is the presence of strong intervening ab- of the survey QSO MUSEUM (Quasar Snapshot Observa- sorption lines of Mg ii  2796, 2803 and Fe ii  2382, tions with MUse: Search for Extended Ultraviolet eMission; 2586 and 2600 at z =1.067. The velocity o set between abs Arrigoni Battaia et al. in prep.). The observations were car- the background galaxies and the intervening absorber is v ried out during UT 2015 September 17 in nominal mode >11,000 km/s. In addition, two emitting galaxies (FG and Downloaded from https://academic.oup.com/mnrasl/advance-article-abstract/doi/10.1093/mnrasl/sly090/5004870 by Ed 'DeepDyve' Gillespie user on 08 June 2018 Spatially Resolved Metal Gas Clouds 3 BGa4). The location of these regions around the central Table 1. Physical properties of the foreground galaxies FG pixel are not unique but driven by SNR considerations to and FG . "b" the impact parameter in kpc, "incl." the inclination result in spectra with similar continuum uxes and hence and "PA" the position angle (East of North).The o set values are with respect to the sky position of the BGa+BGb background homogeneous absorption detection limits. Additionally, the galaxies. The error on the redshift estimates are 0.0001. seeing-limited observations at hand (FWHM=0.74") imply that the spectra in these regions are partially convolved. The masks used for each of these regions are shown in Figure 3. Gal. b z F([O ii]) SFR incl. PA gal Thanks to the remarkable combination of its high-sensitivity [kpc] [erg/s/cm ] [M /yr] [deg] [deg] and IFU capabilities, MUSE observations of this system al- FG 21 1.0677 3.80.410 1.80.7 454 8010 low us to resolve 4 continuous regions within BGa as well FG 63 1.0677 4.10.410 1.90.8 542 743 as the background galaxy BGb thus probing sightlines sep- arated by 1" to 3" (8 to 25 kpc at z ). The o set values are abs measured with reference to the sky position and systemic FG ) are observed in [O ii] at the redshift of the absorber redshift (z =1.06768) of the [O ii] emitter with the lowest abs with angular separations of 21 and 63 kpc respectively (mea- impact parameter, FG . These values are listed in Table 2. sured from the mean of BGa and BGb centroids), well within We measure large rest equivalent widths of Fe ii and their CGM regions. They have R mag=25.7 and 24.5 respec- Mg ii metal lines (Nielsen et al. 2013). With the exception tively. The detection limit at this redshift translates into of the absorber against BGa3, the ratios between the mea- SFR> 0:2M /yr. Figure 1 sketches the system lay-out. sured EWs of Mg ii lines are 1.2 indicating line saturation. We model with Gal t the Sersic pro le of the back- The limited quality of the data however precludes studies ground galaxies determining half-light radii, deconvolved of potential partial coverage. Non-detections are quoted as from the seeing, of R =0.280.6" (BGa) and 0.240.05" e 3- upper limits. The Fe ii lines and Mg ii doublet are not (BGb), corresponding at the redshift of the absorber to detected against the background galaxy BGb (25 kpc-away continuous areas of 17 and 12 kpc respectively. This en- from BGa) down to a signi cant limit of EW<1.4A. We note ables us not only to probe the metal cloud over a total area that there are no indications of these metal absorption lines of 30 kpc but also on scales of 25kpc which is the dis- down to EW<0.7A in the bright quasar 14.3" away (100kpc tance in between the galaxies. To extract the spectrum of at z ). Figure 3 summarises these ndings. The colour map abs each galaxy from the MUSE cube, we used MUSE mpda f indicates the EW values of each region. The corresponding v2.5 (Piqueras et al. 2017). We identi ed pixels associated absorbing spectra are also shown. with each object by running Sextractor on the 2D white Following Ellison et al. (2004) and Rubin et al. (2017), light image. We then extracted the 1D spectrum by inte- we calculate the fractional di erence in EW values with re- grating the ux of the pixels associated with the objects in spect to BGa1. The fractional di erences range from 10{ each wavelength plane. Because the two background galax- 20% (Fe ii  2600) to 0{30% (Mg ii  2796, 2803) in re- ies (separated by 1.5" on the sky) are barely resolved in the gions where the metal lines are detected. Thus, we report MUSE observations and BGb has a faint continuum, we de- only small variations (<30%) on scales of 8kpc (i.e. the ned a level where the pixels in between the two objects inter-regions separation against BGa). However, the non- has lower ux values and extract the spectrum of each ob- detection of Mg ii  2796 against BGb (25 kpc-away) is sig- ject from pixels on either side of this threshold. The [O ii] ni cant and indicates larger variations (>30%) on this scale. emission of the resulting spectra are shown in Figure 2. Therefore, while the data at hand show no indication of sig- The physical properties of the foreground galaxies FG ni cant variations on coherence scales of 8kpc, our ndings and FG are summarised in Table 1. The low impact pa- reveal that the metals traced by cold Mg ii absorptions are rameter galaxy (FG ) is situated 21kpc away from the inhomogeneously distributed on scales smaller than 25 kpc. background galaxies (BGa+BGb), although we cannot ex- We further compute velocity shifts of the absorber pro- clude that some absorption components are related to the les with reference to the systemic redshift of galaxy FG . CGM region of galaxy FG . We measure the [O ii] emission We measure small shifts of the order 3018 km/s with a uxes from a Gaussian t and derive the SFR estimates, possible red and blue component. From a t to the MUSE uncorrected for dust extinction, using the prescription of while light image, we derive the inclination and position an- Kewley et al. (2004). The objects have a SFR of a few solar gle of the foreground galaxies (Table 1). The compactness of masses per year typical of absorbing galaxies observed at FG precludes detailed kinematic analysis (see P eroux et al. these redshifts (P eroux et al. 2011; Rahmani et al. 2016). 2017) but hints at small rotation velocities. The velocity shear observed in absorption could be the signature of the rotation of a gaseous disk extending from the nearby inclined 3 PROPERTIES OF METAL GAS CLOUDS ON galaxy FG (b=21kpc). The small velocities measured could KPC SCALES also be produced by turbulent motions over an area contain- ing several clouds of gas. Given the limited spatial resolution We constrain the physical properties of the metal absorbers of these seeing-limited observations, we cannot disentangle within the background galaxy BGa as well as the BGb which of these scenarios is at play. galaxy. The brightest background object (BGa) is divided The metal cloud size estimate is further combined with in 4 regions separated by 8kpc (BGa1, BGa2, BGa3 and our knowledge of the density to constrain the cloud gas mass. We used the BGa Mg ii 2796 equivalent width relation with H i column density prescription of M enard & Chelouche http://mpdaf.readthedocs.io/en/latest/index.html https://www.astromatic.net/software/sextractor (2009) to estimate the extended gas column density on the Downloaded from https://academic.oup.com/mnrasl/advance-article-abstract/doi/10.1093/mnrasl/sly090/5004870 by Ed 'DeepDyve' Gillespie user on 08 June 2018 4 C. P eroux et al. Figure 2. MUSE observations of the system. Green colours indicate continuum detected objects, while red colours correspond to a pseudo narrow-band lter (7700{7720 A) around [O ii] emission at z =1.067. The spectra on the left show the [O ii] emission lines of abs the background galaxies BGa and BGb at z =1.148. The right panel shows the [O ii] emission lines of the foreground galaxies lying at gal the intervening absorber redshift z =1.067 (FG and FG ). An arbitrarily-scaled sky spectrum is shown in orange. abs Table 2. Spatial variation of the metal absorber physical properties.  is the angular distance in arcsec and "b" the impact parameter in kpc. The o set values and velocity shifts are measured with reference to the sky position and systemic redshift (z =1.0677) of the [O ii] fg emitter with the lowest impact parameter, FG . The redshift measurements are the mean of the Fe ii 2600, Mg ii 2796, 2803 absorption lines measurements, but for BGa4 where only the Mg ii doublet members are used. Non-detections are quoted as 3- upper limits. Galaxies  b SNR z velocity EW EW EW EW EW abs FeII 2382 FeII 2586 FeII 2600 MgII 2796 MgII 2803 ["] [kpc] at Mg ii [km/s] [A] [A] [A] [A] [A] BGa 2.0 16 8.1 ... ... 0.80.2 0.40.2 0.90.2 2.10.3 1.50.2 ...BGa1 2.3 19 4.3 1.06790.0001 +3018 <0.8 <0.8 1.70.4 1.90.3 1.60.4 ...BGa2 0.6 5 4.9 1.06790.0001 +118 <0.7 0.80.4 1.30.5 2.30.6 1.80.5 ...BGa3 1.1 9 5.4 1.06750.0001 3018 <0.6 0.50.3 1.50.4 2.70.4 1.10.4 ...BGa4 2.4 20 3.8 1.06750.0001 2918 <0.9 <0.9 <0.9 1.90.4 1.50.4 BGb 3.1 25 2.5 ... ... <1.4 <1.4 <1.4 <1.4 <1.4 sightline of these two background galaxies. We derive a neu- 20 2 tral gas column density of N (H I )=1.1 10 cm . For sim- plicity we assume the inhomogeneities on 25kpc-scales are due to a spherical cloud, even though the current obser- vations do not rule out asymmetrical geometries (e.g. la- ment). A conservative circular e ective radius <25kpc thus 46 2 corresponds to an area of <1.9 10 cm . Assuming a cov- ering factor of unity, we derive a cold metal mass of <2 10 M . Our data cannot exclude multiple clouds with signi - cantly smaller masses (see e.g. Arrigoni Battaia et al. 2015) thus leading to the estimated upper limit. On the other hand, if the cold gas were to arise from a structure cen- tered towards the opposite direction of BGb, the total mass could be higher. Only high spatial resolution observations of a sample of such systems will be able to address these Figure 3. Physical properties of the metal absorber. The colour issues. Yet, to our knowledge, these measurements are the map indicates the rest-frame equivalent width of Mg ii 2796 in rst direct estimates of the mass of metal cold gas clouds. units of A for each region (see text for de nition). The corre- sponding absorbing spectra are also shown in velocity space. The Fe ii lines and Mg ii doublet are not detected against the back- ground galaxy BGb down to EW<1.4 A, thus providing an upper limit on the size of the cold metal cloud of R<25kpc. 4 DISCUSSION Low-ionisation ions such as Mg ii and Fe ii are typical trac- ers of photo-ionised cold gas in galaxies with temperature within a few hundred kpc of their host galaxies. Simulating T  10 K (Bergeron 1986; Charlton et al. 2003). They are this cold phase of the gas has proven challenging because believed to probe a wide range of neutral hydrogen column of the complexity of the physics involved and because it re- 16 22 2 densities of N (H I )  10 -10 cm (Ellison et al. 2009) quires sub-grid modelling to capture this unresolved physics. Downloaded from https://academic.oup.com/mnrasl/advance-article-abstract/doi/10.1093/mnrasl/sly090/5004870 by Ed 'DeepDyve' Gillespie user on 08 June 2018 Spatially Resolved Metal Gas Clouds 5 However, it is essential to attempt to model the cold gas in Churchill C. W., Vander Vliet J. R., Trujillo-Gomez S., Kacprzak G. G., Klypin A., 2015, ApJ, 802, 10 the CGM to be able to disentangle di erent scenarios and Cooke J., O'Meara J. M., 2015, ApJ, 812, L27 velocity signatures for the absorbers (e.g. disks versus tur- 1 D'Odorico V., et al., 2013, MNRAS, 435, 1198 bulent motion of gas). In "zoom-in" simulations (30 h pc Dedikov S., Shchekinov Y., 2004, Astronomy Reports, 48, 9 resolution at z=0), Churchill et al. (2015) describe these ab- Ellison S. L., Ibata R., Pettini M., Lewis G. F., Aracil B., Petit- sorbers as clouds, i.e. spatially contiguous cells over scales jean P., Srianand R., 2004, A&A, 414, 79 of typically 3 kpc. Hence by studying the Mg ii absorbers Ellison S. L., Murphy M. T., Dessauges-Zavadsky M., 2009, one traces the CGM on kpc scales. MNRAS, 392, 998 Quasar absorbers observations have estimated the char- Hennawi J., et al., 2006, ApJ, 651, 61 acteristic size of gas clouds to be a few dozen of pc. These Husemann B., Bennert V. N., Scharw achter J., Woo J.-H., Choud- results come from indirect photoionisation modelling (i.e. hury O. S., 2016, MNRAS, 455, 1905 Kewley L. J., Geller M. J., Jansen R. A., 2004, AJ, 127, 2002 Werk et al. 2014). Gravitationally lensed quasars more di- Kornei K., Shapley A., Martin C., Coil A., Lotz J., Schiminovich rectly constrain the cloud sizes to be less than  30 pc D., Bundy K., Noeske K., 2012, ApJ, 758, 135 (Rauch et al. 1999; Rubin et al. 2017) but are partially lens- Koyamada S., Misawa T., Inada N., Oguri M., Kashikawa N., model dependent (Lopez et al. 2018). One cannot exclude Okoshi K., 2017, ApJ, 851, 88 that two independent sightlines hit di erent metal clouds Kulkarni V. P., Fall S. M., Lauroesch J. T., York D. G., Welty as expected from density peaks on a single coherent larger D. E., Khare P., Truran J. W., 2005, ApJ, 618, 68 structure, the so-called "blobby sheet" model (Biggs et al. Lopez S., Ellison S., D'Odorico S., Kim 2007, A&A, 469, 61 2016; Koyamada et al. 2017). Lopez S., et al., 2018, Nature, 554, 493 Here we present the serendipitous discovery of extended Martin A., Papastergis E., Giovanelli R., Haynes M., Springob bright background objects with intervening metal absorbers C., Stierwalt S., 2010, ApJ, 723, 1359 Martin C., Shapley A., Coil A., Kornei K., Bundy K., Weiner B., in a seeing-limited MUSE cube. We probe these absorbers Noeske K., Schiminovich D., 2012, ApJ, 760, 127 over a total area of 30 kpc . This test case illustrates the McCourt M., Oh S. P., O'Leary R., Madigan A.-M., 2018, new information now available to characterise the physical MNRAS, 473, 5407 conditions of metal gas clouds. It potentially enables a direct M enard B., Chelouche D., 2009, MNRAS, 393, 808 measure of the spatial distribution, clumpiness and metal Nielsen N., Churchill C., Kacprzak G., 2013, ApJ, 776, 115 cloud sizes which, combined with our knowledge of the den- Oppenheimer B. D., Dav e R., 2006, MNRAS, 373, 1265 sities, constrains the cloud gas masses. In ne, it provides a P eroux C., Bouch e N., Kulkarni V., York D., Vladilo G., 2011, test to the poor-metal mixing scenario by spatially resolving MNRAS, 410, 2237 absorption lines on kpc-scales. P eroux C., et al., 2017, MNRAS, 464, 2053 The spatial resolution of the data is key to achieve these Pettini M., 2003, Cosmochemistry: The Melting Pot of Elements. Cambridge Univ. Press., New York scienti c goals so that adaptive optics GALACSI MUSE Piqueras L., Conseil S., Shepherd M., Bacon R., Leclercq F., observations will play a crucial role by providing seeing- Richard J., 2017, preprint, (arXiv:1710.03554) enhanced observations. In the future, an increased number Quiret S., et al., 2016, MNRAS, 458, 4074 of targets will be within reach thanks to the collecting area Rahmani H., et al., 2016, MNRAS, of the next generation of telescopes combine with IFU ca- Rauch M., Sargent W. L. W., Barlow T. A., 1999, ApJ, 515, 500 pabilities such as the HARMONI instrument (Thatte et al. Rauch M., Sargent W. L. W., Barlow T. A., Carswell R. F., 2001, 2016) under construction for the ELT. At higher redshifts, ApJ, 562, 76 bright targets will be within reach of JWST. Rubin K. H. R., Hennawi J. F., Prochaska J. X., Simcoe R. A., Myers A., Lau M. W., 2015, ApJ, 808, 38 ACKNOWLEDGEMENTS Rubin K. H. R., et al., 2017, preprint, (arXiv:1707.05873) CP is grateful to the ESO and the DFG cluster of excel- Rubin K. H. R., Diamond-Stanic A. M., Coil A. L., Crighton lence `Origin and Structure of the Universe' for support. N. H. M., Moustakas J., 2018, ApJ, 853, 95 RA thanks CNRS and CNES for support for her PhD. Ryan-Weber E. V., Pettini M., Madau P., Zych B. J., 2009, MNRAS, 395, 1476 Schaye J., Aguirre A., 2005, in Hill V., Francois P., Primas F., eds, IAU Symposium Vol. 228, Elemental Tracers of Early Cosmic REFERENCES Evolution. pp 557{568 (arXiv:astro-ph/0509113) Schaye J., Aguirre A., Kim T., Theuns T., Rauch M., Sargent Aguirre A., Hernquist L., Schaye J., Weinberg D. H., Katz N., W., 2003, ApJ, 596, 768 Gardner J., 2001, ApJ, 560, 599 Schaye J., Carswell R. F., Kim T.-S., 2007, MNRAS, 379, 1169 Arrigoni Battaia F., Hennawi J. F., Prochaska J. X., Cantalupo Shull J., Danforth C., Tilton E., 2014, ApJ, 796, 49 S., 2015, ApJ, 809, 163 Thatte N. A., et al., 2016, in Ground-based and Airborne Instru- Bechtold J., Crotts A. P. S., Duncan R. C., Fang Y., 1994, ApJ, mentation for Astronomy VI. p. 99081X 437, L83 Tumlinson J., et al., 2011, ApJ, 733, 111 Bergeron J., 1986, A&A, 155, 8 Weilbacher P., 2015, in Science Operations 2015: ESO/ESA Bergeron J., Boiss e P., 2017, A&A, 604, A37 Workshop, ESO Garching, id.1. p. 1 Biggs A. D., Zwaan M. A., Hatziminaoglou E., P eroux C., Liske Werk J. K., et al., 2014, ApJ, 792, 8 J., 2016, MNRAS, 462, 2819 Bird S., Haehnelt M., Neeleman M., Genel S., Vogelsberger M., Hernquist L., 2015, MNRAS, 447, 1834 Charlton J. C., Ding J., Zonak S. G., Churchill C. W., Bond N. A., Rigby J. R., 2003, ApJ, 589, 111 Chen H.-W., Gauthier J.-R., Sharon K., Johnson S. D., Nair P., Liang C. J., 2014, MNRAS, 438, 1435 Downloaded from https://academic.oup.com/mnrasl/advance-article-abstract/doi/10.1093/mnrasl/sly090/5004870 by Ed 'DeepDyve' Gillespie user on 08 June 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Monthly Notices of the Royal Astronomical Society Letters Oxford University Press

Spatially resolved metal gas clouds

Download PDF
5 pages

Loading next page...
1
 
/lp/ou_press/spatially-resolved-metal-gas-clouds-nmwChm9hAV

References (5)

Publisher
Oxford University Press
Copyright
© 2018 The Author(s) Published by Oxford University Press on behalf of the Royal Astronomical Society
ISSN
1745-3925
eISSN
1745-3933
DOI
10.1093/mnrasl/sly090
Publisher site
See Article on Publisher Site

Abstract

Spatially Resolved Metal Gas Clouds 1 1? 1;2 3 1;3 C. P eroux , H. Rahmani , F. Arrigoni Battaia and R. Augustin Aix Marseille Universit e, CNRS, LAM (Laboratoire d'Astrophysique de Marseille) UMR 7326, 13388, Marseille, France. GEPI, Observatoire de Paris, PSL Universit e, CNRS, 5 Place Jules Janssen, 92190 Meudon, France. European Southern Observatory (ESO), Karl-Schwarzschild-Str.2, D-85748 Garching b. Munc  hen, Germany. Accepted 2018 May 18. Received 2018 May 17; in original form 2018 March 5 ABSTRACT We now have mounting evidences that the circumgalactic medium (CGM) of galaxies is polluted with metals processed through stars. The fate of these metals is however still an open question and several ndings indicate that they remain poorly mixed. A powerful tool to study the low-density gas of the CGM is o ered by absorption lines in quasar spectra, although the information retrieved is limited to 1D along the sightline. We report the serendipitous discovery of two close-by bright z =1.148 extended galaxies with a fortuitous intervening z =1.067 foreground absorber. MUSE gal abs IFU observations spatially probes kpc-scales in absorption in the plane of the sky over a total area spanning 30 kpc . We identify two [O ii] emitters at z down to 21 kpc with SFR2 M /yr. abs We measure small fractional variations (<30%) in the equivalent widths of Fe ii and Mg ii cold gas absorbers on coherence scales of 8kpc but stronger variation on larger scales (25kpc). We compute the corresponding cloud gas mass <2 10 M . Our results indicate a good eciency of the metal mixing on kpc-scales in the CGM of a typical z1 galaxy. This study show-cases new prospects for mapping the distribution and sizes of metal clouds observed in absorption against extended background sources with 3D spectroscopy. Key words: galaxies: ISM { quasars: absorption lines { intergalactic medium 1 INTRODUCTION also Bird et al. 2015). These authors predict a mismatch be- tween the smoothing scales of H i and high-ionisation metals Baryons from the cosmic web are known to accrete ef- which has yet to be witnessed with observations. These nd- ciently onto galaxies. This mechanism sustains violent ings question how traditional absorption lines studies recover episodes of star formation which power out ows extending the true gas properties. Based on hydrodynamical equilib- out to the surrounding circumgalactic medium (CGM), and rium arguments, McCourt et al. (2018) further suggest that even reaching the larger scales of the intergalactic medium the CGM of galaxies cools via "shattering", resulting in a or IGM (Aguirre et al. 2001; Oppenheimer & Dav e 2006). high covering fraction of pc-scale, photoionised cloudlets. The metals carried by the out ows will either rain back Important issues thus remain unsolved: On which scales on to galaxies or get mixed into the IGM. Indeed, obser- are metals mixed? How does it vary with environment (i.e. vations of the IGM indicate signi cant quantities of met- IGM vs. CGM)? Do low and high-ionisation ions have dif- als at all redshifts (Pettini 2003; Ryan-Weber et al. 2009; ferent coherence scales? Can we nd direct observational D'Odorico et al. 2013; Shull et al. 2014). However, the mix- evidence of shattering on pc-scales? Where are the high- ing could remain incomplete (e.g. Dedikov & Shchekinov metallicity, intergalactic gas clouds? If the H i absorption 2004). Speci cally, Schaye et al. (2007) show that ionised and high-ionisation metal absorption gas arise in distinct metal clouds are compact (typical scales of 100pc) and physical gas structures, the observational techniques em- are short-lived. Once in the IGM they expand until they ployed to infer metallicities and the total mass of the warm- reach pressure equilibrium with their environment but re- hot CGM gas would be challenged (Tumlinson et al. 2011; main poorly mixed on scales of 1kpc or smaller. More re- Werk et al. 2014). The pockets of metal-rich material have cently, Churchill et al. (2015) nd that in the CGM of a also profound implications on metal-cooling eciency and in simulated dwarf galaxy, low ionization gas arises cloud struc- turn galaxy formation. Our understanding of these phases tures of scales of order  3kpc. High ionization gas how- of the gas and their metallicities has so far been limited by ever lies in multiple extended structures spread over 100 the lack of observational constraints. kpc and due to complex velocity elds, highly separated A powerful tool to study this low-density gas is o ered structures give rise to absorption at similar velocities (see by absorption lines in quasar spectra. In these quasar ab- sorbers, the minimum gas density that can be detected is E-mail:celine.peroux@gmail.com set by the brightness of the background source and thus Downloaded from https://academic.oup.com/mnrasl/advance-article-abstract/doi/10.1093/mnrasl/sly090/5004870 by Ed 'DeepDyve' Gillespie user on 08 June 2018 2 C. P eroux et al. the detection eciency is independent of redshift. High- zQSO=3.33 z =1.15 BG Background Quasar Background Galaxies quality quasar absorption spectra have produced a wealth of information regarding the distribution of heavy ele- ments (e.g. Kulkarni et al. 2005; Quiret et al. 2016). How- ever, the brightest background sources (quasars and gamma- ray bursts) are point-sources so that the observer is lim- ited to the information gained along the line-of-sight. The metallicity we typically infer from absorption studies is then zFG=1.07 not determined by the abundances of heavy elements on Foreground Galaxies the size of the metal concentrations, but by the metallic- zabs=1.07 ity smoothed over the size of the HI absorber, which is absorbers in typically 100 kpc (Bechtold et al. 1994; Schaye et al. 2003; Background Galaxy spectra Schaye & Aguirre 2005). On smaller scales the distribu- tion of metals is essentially unknown. To remedy this, ob- Figure 1. Sketch of the system lay-out. A bright z =3.33 quasar qso servers have used close quasar pairs (Hennawi et al. 2006; was the original target of the observations. In these data, we have Martin et al. 2010; Rubin et al. 2015, 2018) as well as mul- serendipitously discovered an intervening z =1.067 metal ab- abs tiple images from gravitationally lensed background objects sorber in the spectrum of two close-by z =1.148 extended back- gal to probe the transverse small-scale coherence along lines-of- ground galaxies. In addition, two emitting galaxies are observed sight dozen kpc apart (Rauch et al. 2001; Ellison et al. 2004; in [O ii] at the redshift of the absorber. Lopez et al. 2007; Chen et al. 2014; Rubin et al. 2017). By using an extended galaxy as background source however, one can directly map the distribution and sizes and consisted of three exposures of 900 s each. The sub- of the metal absorbers on small scales. By using this set- exposures were rotated by 90 degrees to minimise residuals up, Cooke & O'Meara (2015) have estimated that indeed from uneven at- elding. The eld of view is 60 "  60 ", high-column density neutral gas can span continuous areas a 0.2 "/pixel scale and a spectral sampling of 1.25 A/pixel 8 10 10 {10 times larger than previously explored in quasar or covering 4750{9350A. gamma-ray burst sightlines. Bergeron & Boiss e (2017) have The data were reduced with version v1.6.4 of the ESO used intervening absorber lines redshifted on more extended MUSE pipeline (Weilbacher 2015) and additional external quasar emission lines to probe the spatial covering of the gas routines for sky subtraction as explained in the following. clouds. In a remarkable work, Lopez et al. (2018) reported Master bias, at eld images and arc lamp exposures were Mg ii absorption along a bright lensed arc probing scales of produced based on data taken closest in time to the sci- the order 2{4kpc. They nd that the strength of the absorber ence frames. The raw science and standard star cubes were decreases with radius from the emitting galaxy as expected then processed correcting the wavelength calibration to a from the quasar absorber population. However the physical heliocentric reference. We checked the wavelength solution properties derived from the observations of lensed systems using the known wavelengths of the night-sky OH lines and rely heavily on the lensed model to compute the magni ca- nd it to be accurate within 18 km/s. The individual expo- tion factors and survey di erent physical area in the image sures were registered using the central quasar to ensure ac- plane, thus probing inhomogeneous ux levels. curate relative astrometry. Finally, the individual exposures Here, we report the serendipitous discovery of two were combined into a single data cube. The removal of OH bright z =1.148 extended galaxies with a fortuitous inter- gal emission lines from the night sky was accomplished with an vening foreground absorber at z =1.067 along their sight- abs additional purpose-developed code tested in previous work lines. The manuscript is organised as follows: Section 2 pro- (P eroux et al. 2017). After selecting sky regions in the eld, vides the observational set-up and details of the system lay- we created Principle Component Analysis (PCA) compo- out. Section 3 presents the constraints on the physical con- nents from the spectra which were further applied to the sci- ditions of the metal gas clouds. Finally, in section 4, we ence datacube to remove sky line residuals (Husemann et al. review the impact of these ndings in the broader context 2016, trouble). The seeing of the nal combined data was of the metal mixing in the CGM of galaxies. Throughout measured from the quasar. The resulting point spread func- 1 1 this paper we adopt an H = 70 km s Mpc , = 0:3, 0 M tion (PSF) has a full width at half maximum of 0:74" at the and = 0:7 cosmology. At the redshift of the absorber wavelength of the Mg ii absorber (5780 A), corresponding (z =1.067), 1" corresponds to 8.2 kpc. abs to 6 kpc at z =1.067 or 0:68" at 7700A. abs In this MUSE cube, we report the serendipitous dis- covery of two nearby z =1.148 [O ii] emitters, coined BGa gal and BGb, with bright continua (R mag=21.9 and 23.3 re- 2 MUSE OBSERVATIONS OF A spectively). The spectra of these background galaxies show REMARKABLE SYSTEM evidences of a "down-the-barrel" out ow with strong absorp- MUSE observations of the eld of quasar SDSS J02500757 tions in Mg ii, Mg i and Fe ii typical of bright galaxies at z=1 were undertaken in service mode in natural seeing mode un- (Kornei et al. 2012; Martin et al. 2012). A more remarkable der programme 095.A-0615(A). These observations are part feature however is the presence of strong intervening ab- of the survey QSO MUSEUM (Quasar Snapshot Observa- sorption lines of Mg ii  2796, 2803 and Fe ii  2382, tions with MUse: Search for Extended Ultraviolet eMission; 2586 and 2600 at z =1.067. The velocity o set between abs Arrigoni Battaia et al. in prep.). The observations were car- the background galaxies and the intervening absorber is v ried out during UT 2015 September 17 in nominal mode >11,000 km/s. In addition, two emitting galaxies (FG and Downloaded from https://academic.oup.com/mnrasl/advance-article-abstract/doi/10.1093/mnrasl/sly090/5004870 by Ed 'DeepDyve' Gillespie user on 08 June 2018 Spatially Resolved Metal Gas Clouds 3 BGa4). The location of these regions around the central Table 1. Physical properties of the foreground galaxies FG pixel are not unique but driven by SNR considerations to and FG . "b" the impact parameter in kpc, "incl." the inclination result in spectra with similar continuum uxes and hence and "PA" the position angle (East of North).The o set values are with respect to the sky position of the BGa+BGb background homogeneous absorption detection limits. Additionally, the galaxies. The error on the redshift estimates are 0.0001. seeing-limited observations at hand (FWHM=0.74") imply that the spectra in these regions are partially convolved. The masks used for each of these regions are shown in Figure 3. Gal. b z F([O ii]) SFR incl. PA gal Thanks to the remarkable combination of its high-sensitivity [kpc] [erg/s/cm ] [M /yr] [deg] [deg] and IFU capabilities, MUSE observations of this system al- FG 21 1.0677 3.80.410 1.80.7 454 8010 low us to resolve 4 continuous regions within BGa as well FG 63 1.0677 4.10.410 1.90.8 542 743 as the background galaxy BGb thus probing sightlines sep- arated by 1" to 3" (8 to 25 kpc at z ). The o set values are abs measured with reference to the sky position and systemic FG ) are observed in [O ii] at the redshift of the absorber redshift (z =1.06768) of the [O ii] emitter with the lowest abs with angular separations of 21 and 63 kpc respectively (mea- impact parameter, FG . These values are listed in Table 2. sured from the mean of BGa and BGb centroids), well within We measure large rest equivalent widths of Fe ii and their CGM regions. They have R mag=25.7 and 24.5 respec- Mg ii metal lines (Nielsen et al. 2013). With the exception tively. The detection limit at this redshift translates into of the absorber against BGa3, the ratios between the mea- SFR> 0:2M /yr. Figure 1 sketches the system lay-out. sured EWs of Mg ii lines are 1.2 indicating line saturation. We model with Gal t the Sersic pro le of the back- The limited quality of the data however precludes studies ground galaxies determining half-light radii, deconvolved of potential partial coverage. Non-detections are quoted as from the seeing, of R =0.280.6" (BGa) and 0.240.05" e 3- upper limits. The Fe ii lines and Mg ii doublet are not (BGb), corresponding at the redshift of the absorber to detected against the background galaxy BGb (25 kpc-away continuous areas of 17 and 12 kpc respectively. This en- from BGa) down to a signi cant limit of EW<1.4A. We note ables us not only to probe the metal cloud over a total area that there are no indications of these metal absorption lines of 30 kpc but also on scales of 25kpc which is the dis- down to EW<0.7A in the bright quasar 14.3" away (100kpc tance in between the galaxies. To extract the spectrum of at z ). Figure 3 summarises these ndings. The colour map abs each galaxy from the MUSE cube, we used MUSE mpda f indicates the EW values of each region. The corresponding v2.5 (Piqueras et al. 2017). We identi ed pixels associated absorbing spectra are also shown. with each object by running Sextractor on the 2D white Following Ellison et al. (2004) and Rubin et al. (2017), light image. We then extracted the 1D spectrum by inte- we calculate the fractional di erence in EW values with re- grating the ux of the pixels associated with the objects in spect to BGa1. The fractional di erences range from 10{ each wavelength plane. Because the two background galax- 20% (Fe ii  2600) to 0{30% (Mg ii  2796, 2803) in re- ies (separated by 1.5" on the sky) are barely resolved in the gions where the metal lines are detected. Thus, we report MUSE observations and BGb has a faint continuum, we de- only small variations (<30%) on scales of 8kpc (i.e. the ned a level where the pixels in between the two objects inter-regions separation against BGa). However, the non- has lower ux values and extract the spectrum of each ob- detection of Mg ii  2796 against BGb (25 kpc-away) is sig- ject from pixels on either side of this threshold. The [O ii] ni cant and indicates larger variations (>30%) on this scale. emission of the resulting spectra are shown in Figure 2. Therefore, while the data at hand show no indication of sig- The physical properties of the foreground galaxies FG ni cant variations on coherence scales of 8kpc, our ndings and FG are summarised in Table 1. The low impact pa- reveal that the metals traced by cold Mg ii absorptions are rameter galaxy (FG ) is situated 21kpc away from the inhomogeneously distributed on scales smaller than 25 kpc. background galaxies (BGa+BGb), although we cannot ex- We further compute velocity shifts of the absorber pro- clude that some absorption components are related to the les with reference to the systemic redshift of galaxy FG . CGM region of galaxy FG . We measure the [O ii] emission We measure small shifts of the order 3018 km/s with a uxes from a Gaussian t and derive the SFR estimates, possible red and blue component. From a t to the MUSE uncorrected for dust extinction, using the prescription of while light image, we derive the inclination and position an- Kewley et al. (2004). The objects have a SFR of a few solar gle of the foreground galaxies (Table 1). The compactness of masses per year typical of absorbing galaxies observed at FG precludes detailed kinematic analysis (see P eroux et al. these redshifts (P eroux et al. 2011; Rahmani et al. 2016). 2017) but hints at small rotation velocities. The velocity shear observed in absorption could be the signature of the rotation of a gaseous disk extending from the nearby inclined 3 PROPERTIES OF METAL GAS CLOUDS ON galaxy FG (b=21kpc). The small velocities measured could KPC SCALES also be produced by turbulent motions over an area contain- ing several clouds of gas. Given the limited spatial resolution We constrain the physical properties of the metal absorbers of these seeing-limited observations, we cannot disentangle within the background galaxy BGa as well as the BGb which of these scenarios is at play. galaxy. The brightest background object (BGa) is divided The metal cloud size estimate is further combined with in 4 regions separated by 8kpc (BGa1, BGa2, BGa3 and our knowledge of the density to constrain the cloud gas mass. We used the BGa Mg ii 2796 equivalent width relation with H i column density prescription of M enard & Chelouche http://mpdaf.readthedocs.io/en/latest/index.html https://www.astromatic.net/software/sextractor (2009) to estimate the extended gas column density on the Downloaded from https://academic.oup.com/mnrasl/advance-article-abstract/doi/10.1093/mnrasl/sly090/5004870 by Ed 'DeepDyve' Gillespie user on 08 June 2018 4 C. P eroux et al. Figure 2. MUSE observations of the system. Green colours indicate continuum detected objects, while red colours correspond to a pseudo narrow-band lter (7700{7720 A) around [O ii] emission at z =1.067. The spectra on the left show the [O ii] emission lines of abs the background galaxies BGa and BGb at z =1.148. The right panel shows the [O ii] emission lines of the foreground galaxies lying at gal the intervening absorber redshift z =1.067 (FG and FG ). An arbitrarily-scaled sky spectrum is shown in orange. abs Table 2. Spatial variation of the metal absorber physical properties.  is the angular distance in arcsec and "b" the impact parameter in kpc. The o set values and velocity shifts are measured with reference to the sky position and systemic redshift (z =1.0677) of the [O ii] fg emitter with the lowest impact parameter, FG . The redshift measurements are the mean of the Fe ii 2600, Mg ii 2796, 2803 absorption lines measurements, but for BGa4 where only the Mg ii doublet members are used. Non-detections are quoted as 3- upper limits. Galaxies  b SNR z velocity EW EW EW EW EW abs FeII 2382 FeII 2586 FeII 2600 MgII 2796 MgII 2803 ["] [kpc] at Mg ii [km/s] [A] [A] [A] [A] [A] BGa 2.0 16 8.1 ... ... 0.80.2 0.40.2 0.90.2 2.10.3 1.50.2 ...BGa1 2.3 19 4.3 1.06790.0001 +3018 <0.8 <0.8 1.70.4 1.90.3 1.60.4 ...BGa2 0.6 5 4.9 1.06790.0001 +118 <0.7 0.80.4 1.30.5 2.30.6 1.80.5 ...BGa3 1.1 9 5.4 1.06750.0001 3018 <0.6 0.50.3 1.50.4 2.70.4 1.10.4 ...BGa4 2.4 20 3.8 1.06750.0001 2918 <0.9 <0.9 <0.9 1.90.4 1.50.4 BGb 3.1 25 2.5 ... ... <1.4 <1.4 <1.4 <1.4 <1.4 sightline of these two background galaxies. We derive a neu- 20 2 tral gas column density of N (H I )=1.1 10 cm . For sim- plicity we assume the inhomogeneities on 25kpc-scales are due to a spherical cloud, even though the current obser- vations do not rule out asymmetrical geometries (e.g. la- ment). A conservative circular e ective radius <25kpc thus 46 2 corresponds to an area of <1.9 10 cm . Assuming a cov- ering factor of unity, we derive a cold metal mass of <2 10 M . Our data cannot exclude multiple clouds with signi - cantly smaller masses (see e.g. Arrigoni Battaia et al. 2015) thus leading to the estimated upper limit. On the other hand, if the cold gas were to arise from a structure cen- tered towards the opposite direction of BGb, the total mass could be higher. Only high spatial resolution observations of a sample of such systems will be able to address these Figure 3. Physical properties of the metal absorber. The colour issues. Yet, to our knowledge, these measurements are the map indicates the rest-frame equivalent width of Mg ii 2796 in rst direct estimates of the mass of metal cold gas clouds. units of A for each region (see text for de nition). The corre- sponding absorbing spectra are also shown in velocity space. The Fe ii lines and Mg ii doublet are not detected against the back- ground galaxy BGb down to EW<1.4 A, thus providing an upper limit on the size of the cold metal cloud of R<25kpc. 4 DISCUSSION Low-ionisation ions such as Mg ii and Fe ii are typical trac- ers of photo-ionised cold gas in galaxies with temperature within a few hundred kpc of their host galaxies. Simulating T  10 K (Bergeron 1986; Charlton et al. 2003). They are this cold phase of the gas has proven challenging because believed to probe a wide range of neutral hydrogen column of the complexity of the physics involved and because it re- 16 22 2 densities of N (H I )  10 -10 cm (Ellison et al. 2009) quires sub-grid modelling to capture this unresolved physics. Downloaded from https://academic.oup.com/mnrasl/advance-article-abstract/doi/10.1093/mnrasl/sly090/5004870 by Ed 'DeepDyve' Gillespie user on 08 June 2018 Spatially Resolved Metal Gas Clouds 5 However, it is essential to attempt to model the cold gas in Churchill C. W., Vander Vliet J. R., Trujillo-Gomez S., Kacprzak G. G., Klypin A., 2015, ApJ, 802, 10 the CGM to be able to disentangle di erent scenarios and Cooke J., O'Meara J. M., 2015, ApJ, 812, L27 velocity signatures for the absorbers (e.g. disks versus tur- 1 D'Odorico V., et al., 2013, MNRAS, 435, 1198 bulent motion of gas). In "zoom-in" simulations (30 h pc Dedikov S., Shchekinov Y., 2004, Astronomy Reports, 48, 9 resolution at z=0), Churchill et al. (2015) describe these ab- Ellison S. L., Ibata R., Pettini M., Lewis G. F., Aracil B., Petit- sorbers as clouds, i.e. spatially contiguous cells over scales jean P., Srianand R., 2004, A&A, 414, 79 of typically 3 kpc. Hence by studying the Mg ii absorbers Ellison S. L., Murphy M. T., Dessauges-Zavadsky M., 2009, one traces the CGM on kpc scales. MNRAS, 392, 998 Quasar absorbers observations have estimated the char- Hennawi J., et al., 2006, ApJ, 651, 61 acteristic size of gas clouds to be a few dozen of pc. These Husemann B., Bennert V. N., Scharw achter J., Woo J.-H., Choud- results come from indirect photoionisation modelling (i.e. hury O. S., 2016, MNRAS, 455, 1905 Kewley L. J., Geller M. J., Jansen R. A., 2004, AJ, 127, 2002 Werk et al. 2014). Gravitationally lensed quasars more di- Kornei K., Shapley A., Martin C., Coil A., Lotz J., Schiminovich rectly constrain the cloud sizes to be less than  30 pc D., Bundy K., Noeske K., 2012, ApJ, 758, 135 (Rauch et al. 1999; Rubin et al. 2017) but are partially lens- Koyamada S., Misawa T., Inada N., Oguri M., Kashikawa N., model dependent (Lopez et al. 2018). One cannot exclude Okoshi K., 2017, ApJ, 851, 88 that two independent sightlines hit di erent metal clouds Kulkarni V. P., Fall S. M., Lauroesch J. T., York D. G., Welty as expected from density peaks on a single coherent larger D. E., Khare P., Truran J. W., 2005, ApJ, 618, 68 structure, the so-called "blobby sheet" model (Biggs et al. Lopez S., Ellison S., D'Odorico S., Kim 2007, A&A, 469, 61 2016; Koyamada et al. 2017). Lopez S., et al., 2018, Nature, 554, 493 Here we present the serendipitous discovery of extended Martin A., Papastergis E., Giovanelli R., Haynes M., Springob bright background objects with intervening metal absorbers C., Stierwalt S., 2010, ApJ, 723, 1359 Martin C., Shapley A., Coil A., Kornei K., Bundy K., Weiner B., in a seeing-limited MUSE cube. We probe these absorbers Noeske K., Schiminovich D., 2012, ApJ, 760, 127 over a total area of 30 kpc . This test case illustrates the McCourt M., Oh S. P., O'Leary R., Madigan A.-M., 2018, new information now available to characterise the physical MNRAS, 473, 5407 conditions of metal gas clouds. It potentially enables a direct M enard B., Chelouche D., 2009, MNRAS, 393, 808 measure of the spatial distribution, clumpiness and metal Nielsen N., Churchill C., Kacprzak G., 2013, ApJ, 776, 115 cloud sizes which, combined with our knowledge of the den- Oppenheimer B. D., Dav e R., 2006, MNRAS, 373, 1265 sities, constrains the cloud gas masses. In ne, it provides a P eroux C., Bouch e N., Kulkarni V., York D., Vladilo G., 2011, test to the poor-metal mixing scenario by spatially resolving MNRAS, 410, 2237 absorption lines on kpc-scales. P eroux C., et al., 2017, MNRAS, 464, 2053 The spatial resolution of the data is key to achieve these Pettini M., 2003, Cosmochemistry: The Melting Pot of Elements. Cambridge Univ. Press., New York scienti c goals so that adaptive optics GALACSI MUSE Piqueras L., Conseil S., Shepherd M., Bacon R., Leclercq F., observations will play a crucial role by providing seeing- Richard J., 2017, preprint, (arXiv:1710.03554) enhanced observations. In the future, an increased number Quiret S., et al., 2016, MNRAS, 458, 4074 of targets will be within reach thanks to the collecting area Rahmani H., et al., 2016, MNRAS, of the next generation of telescopes combine with IFU ca- Rauch M., Sargent W. L. W., Barlow T. A., 1999, ApJ, 515, 500 pabilities such as the HARMONI instrument (Thatte et al. Rauch M., Sargent W. L. W., Barlow T. A., Carswell R. F., 2001, 2016) under construction for the ELT. At higher redshifts, ApJ, 562, 76 bright targets will be within reach of JWST. Rubin K. H. R., Hennawi J. F., Prochaska J. X., Simcoe R. A., Myers A., Lau M. W., 2015, ApJ, 808, 38 ACKNOWLEDGEMENTS Rubin K. H. R., et al., 2017, preprint, (arXiv:1707.05873) CP is grateful to the ESO and the DFG cluster of excel- Rubin K. H. R., Diamond-Stanic A. M., Coil A. L., Crighton lence `Origin and Structure of the Universe' for support. N. H. M., Moustakas J., 2018, ApJ, 853, 95 RA thanks CNRS and CNES for support for her PhD. Ryan-Weber E. V., Pettini M., Madau P., Zych B. J., 2009, MNRAS, 395, 1476 Schaye J., Aguirre A., 2005, in Hill V., Francois P., Primas F., eds, IAU Symposium Vol. 228, Elemental Tracers of Early Cosmic REFERENCES Evolution. pp 557{568 (arXiv:astro-ph/0509113) Schaye J., Aguirre A., Kim T., Theuns T., Rauch M., Sargent Aguirre A., Hernquist L., Schaye J., Weinberg D. H., Katz N., W., 2003, ApJ, 596, 768 Gardner J., 2001, ApJ, 560, 599 Schaye J., Carswell R. F., Kim T.-S., 2007, MNRAS, 379, 1169 Arrigoni Battaia F., Hennawi J. F., Prochaska J. X., Cantalupo Shull J., Danforth C., Tilton E., 2014, ApJ, 796, 49 S., 2015, ApJ, 809, 163 Thatte N. A., et al., 2016, in Ground-based and Airborne Instru- Bechtold J., Crotts A. P. S., Duncan R. C., Fang Y., 1994, ApJ, mentation for Astronomy VI. p. 99081X 437, L83 Tumlinson J., et al., 2011, ApJ, 733, 111 Bergeron J., 1986, A&A, 155, 8 Weilbacher P., 2015, in Science Operations 2015: ESO/ESA Bergeron J., Boiss e P., 2017, A&A, 604, A37 Workshop, ESO Garching, id.1. p. 1 Biggs A. D., Zwaan M. A., Hatziminaoglou E., P eroux C., Liske Werk J. K., et al., 2014, ApJ, 792, 8 J., 2016, MNRAS, 462, 2819 Bird S., Haehnelt M., Neeleman M., Genel S., Vogelsberger M., Hernquist L., 2015, MNRAS, 447, 1834 Charlton J. C., Ding J., Zonak S. G., Churchill C. W., Bond N. A., Rigby J. R., 2003, ApJ, 589, 111 Chen H.-W., Gauthier J.-R., Sharon K., Johnson S. D., Nair P., Liang C. J., 2014, MNRAS, 438, 1435 Downloaded from https://academic.oup.com/mnrasl/advance-article-abstract/doi/10.1093/mnrasl/sly090/5004870 by Ed 'DeepDyve' Gillespie user on 08 June 2018

Journal

Monthly Notices of the Royal Astronomical Society Letters – Oxford University Press

Published: Jun 25, 2018

Keywords: intergalactic medium; galaxies: ISM; quasars: absorption lines

There are no references for this article.