A translucent interstellar cloud at z=2.69: CO, H2 and HD in the line-of-sight to SDSS J123714.60+064759.5
P. Noterdaeme, P. Petitjean, C. Ledoux, S. Lopez, R. Srianand, S.D. Vergani
aa r X i v : . [ a s t r o - ph . C O ] A ug Astronomy&Astrophysicsmanuscript no. 1237 c (cid:13)
ESO 2018June 7, 2018
A translucent interstellar cloud at z = . : ⋆ CO, H and HD in the line-of-sight to SDSS J123714.60 + P. Noterdaeme , P. Petitjean , C. Ledoux , S. Lopez , R. Srianand and S. D. Vergani , Departamento de Astronom´ıa, Universidad de Chile, Casilla 36-D, Santiago, Chilee-mail: [email protected], [email protected] Universit´e Paris 6, Institut d’Astrophysique de Paris, CNRS UMR 7095, 98bis bd Arago, 75014 Paris, Francee-mail: [email protected] European Southern Observatory, Alonso de C´ordova 3107, Vitacura, Casilla 19001, Santiago 19, Chilee-mail: [email protected] Inter-University Centre for Astronomy and Astrophysics, Post Bag 4, Ganeshkhind, 411 007 Pune, Indiae-mail: [email protected] Universit´e Paris 7, APC, CNRS UMR 7164, 10 rue Alice Domon et L´eonie Duquet, 75205 Paris Cedex 13, France GEPI, Observatoire de Paris, CNRS UMR 8111, 5 place Jules Janssen, 92195 Meudon, Francee-mail: [email protected]
Received / Accepted
ABSTRACT
We present the analysis of a sub-Damped Lyman- α system with neutral hydrogen column density, log N (H ) (cm − ) = ± z abs = + z em = / UVES and X-shooter spectrographs, we detect H ,HD and CO molecules in absorption with log N (H , HD, CO) (cm − ) = + . − . , 14.48 ± ± / H] = + / Zn] = − . velocity cxomponents spanning ∼
125 km s − are detected. The strongest H component,at z abs = N (H ) = i absorption. This implies that the molecularfraction in this component, f H2 = N (H ) / (2 N (H ) + N (H )), is larger than the mean molecular fraction h f H2 i = / associated with this H component having N (Cl ) / N (Cl + ) > is tied up to H bycharge exchange reactions, this means that the molecular fraction in this component is not far from unity. The kinetic temperaturederived from the J = is T = + − K and the particle density derived from the C ground-state finestructure level populations is n H0 ∼ − . We derive an electronic density < − for a UV field similar to the Galactic one andshow that the carbon to sulphur ratio in the cloud is close to the solar ratio. The size of the molecular cloud is probably smaller than1 pc. Both the CO / H = − and CO / C ∼ f H2 > A V = .
14, albeit lower than the definition of atranslucent sightline (based on extinction properties), is high for the observed H column density. This means that intervening cloudswith similar local properties but with larger column densities (i.e. larger physical extent) could be missed by current magnitude-limitedQSO surveys. The excitation of CO is dominated by radiative interaction with the Cosmic Microwave Background Radiation (CMBR)and we derive T ex (CO) = + . − . K when T CMBR ( z = = N (HD) / N (H ) = − . This is about10 times higher than what is measured in the Galactic ISM for f H2 = / / H ratio – is only about 3. This can bethe consequence of accretion of unprocessed gas from the intergalactic medium onto the associated galaxy. In the future, it will bepossible to search e ffi ciently for molecular-rich DLAs / sub-DLAs with X-shooter but detailed studies of the physical state of the gaswill still need UVES observations. Key words.
Cosmology: Observations - Galaxies: ISM - Quasars: Absorption lines - Quasars: Individual:SDSS J123714.60 +
1. Introduction
Studies of the Interstellar Medium (ISM) in the local Universehave shown that the neutral ISM presents a complex struc-ture, with cold and dense clouds immersed in a warmer andmore di ff use medium. These di ff erent ISM phases should be Send o ff print requests to : P. Noterdaeme ⋆ Based on observations carried out with X-shooter and theUltraviolet and Visual Echelle Spectrograph (UVES), both mountedon the European Southern Observatory Very Large Telescope Unit 2- Kueyen, under Program IDs 082.A-0544(A), 083.A-0454(A) and084.A-0699(A). detectable at high redshift by their absorption signatures inDamped Lyman- α (DLA) systems observed in quasar spectra(Petitjean et al., 1992). However, although there are evidences ofthe multiphase nature of DLA systems (e.g. Wolfe et al., 2004),most of the intervening DLAs probe only warm ( T > ∼ ff use ( n H < − ) atomic gas (e.g. Petitjean et al.,2000; Kanekar & Chengalur, 2003). The reason is that the cross-sections of the di ff erent phases are quite di ff erent and it is notpossible to sample them equally well.Searching for molecular hydrogen in high redshift DLAs(Ledoux et al., 2003; Petitjean et al., 2006; Noterdaeme et al.,2008a) is an e ffi cient way of detecting colder and denser
1. Noterdaeme et al.: A translucent interstellar cloud at z = . neutral gas and to probe its physical conditions (e.g.Reimers et al., 2003; Cui et al., 2005; Hirashita & Ferrara, 2005;Srianand et al., 2005; Ledoux et al., 2006b; Noterdaeme et al.,2007a,b). These studies have shown that molecular hydrogenis confined in small clouds (pc-sized) with densities n ∼ − and temperatures T ∼ -bearing clouds in DLAs is much less than one andonly 10 % of the lines of sight through a DLA galaxy do inter-cept H -bearing clouds down to a limit of N (H ) ∼ cm − (Noterdaeme et al., 2008a). H -bearing clouds in DLAs havesmall physical extents. Direct evidence for this is given bythe fact that the intervening H -bearing gas does not com-pletely fill the beam from the broad line region of the quasarQ 1232 +
082 (Ivanchik et al. 2010; Balashev et al., submitted).Nonetheless, the molecular fraction in DLAs remains small andtypical of what is seen in Galactic di ff use atomic gas with f H2 = N (H ) / (2 N (H ) + N (H )) < . ff useatomic, with lowmolecular fractions; (ii) di ff usemolecular, where the fraction ofhydrogen in molecules becomes substantial ( f H2 > .
1) but car-bon is still mainly in ionised form (C + ); (iii) translucent (firstintroduced by van Dishoeck & Black, 1989), where the carbonmakes the transition to molecular; and (iv) dense molecular,where both hydrogen and carbon are fully molecular. As dis-cussed above, most of the H -bearing DLAs detected so far arepart of the first, and maybe for some of them, part of the secondcategories. The fourth category may be di ffi cult to detect in ab-sorption because of the high extinction such a cloud produces onthe background source.Despite their highly interesting chemistry and their closeconnection with star formation, we know very little abouttranslucent clouds (i.e., the third category) at high redshift. Thesmall cross-section of these clouds and / or the induced extinc-tion of the light from the background sources can probably ex-plain the absence of detection in more than three decades of QSOabsorption-line research. However, observing molecular-rich gasin absorption should be possible by selecting sightlines passingthrough or starting from star-forming regions.Since long-duration Gamma-Ray Bursts (GRBs) are knownto occur within star-forming regions, absorption lines at thehost-galaxy redshifts which are imprinted in GRB optical af-terglow spectra (e.g., Fynbo et al., 2009) are obvious targets to-wards this goal. Nevertheless, current samples are characterisedby a general lack of H detection (e.g., Fynbo et al., 2006;Tumlinson et al., 2007). This is probably due to the still limitedsample sizes as well as a bias against dusty – molecular-rich –lines of sight (Ledoux et al., 2009; Fynbo et al., 2009). The firstdetection of both H and CO in the low-resolution spectrum of ahighly reddened GRB afterglow (Prochaska et al., 2009) seemsto confirm this scenario. Moreover, the observed molecular ex-citation is high in this case, indicating strong UV pumping fromthe GRB afterglow itself.With the large number of quasar spectra available in theSloan Digital Sky Survey (SDSS), it becomes possible to se-lect the rare sightlines passing through intervening molecular-rich gas. However, due to the small cross-section of such clouds,an e ffi cient selection must be applied. In the local ISM, car-bon is found to transition from a ionised state (C + ) to neu-tral (C ) and molecular form (CO) from the most superficialto the deepest parts of the clouds (e.g. Snow & McCall, 2006;Burgh et al., 2010). From our Very Large Telescope (VLT) sur-vey for H in DLAs (Ledoux et al., 2003; Noterdaeme et al., 2008a), it appears that C is generally observed in the same com-ponents as H . This is due to the photo-ionisation potential ofC being similar to the energy of photons that dissociate H .However, the neutral fraction of carbon is generally small, prob-ably because the gas is not completely shielded. This explainsthe non-detection of CO in these H -bearing DLAs, even downto N (CO) ∼ cm − (e.g. Petitjean et al., 2002). Searchingfor systems with large column densities of neutral carbon couldbe an e ffi cient way to select more shielded gas where othermolecules can survive, without relying on a pre-selection basedon the H column density (i.e. the absorbers need not be DLAs).Since several C i lines are located redwards of the Lyman- α for-est, it is possible to search for strong C i absorptions directly inSDSS spectra using automatic procedures. We therefore initi-ated a program to survey with the VLT such specific sightlines.Our selection has been very successful and already allowed usto detect carbon monoxide along QSO sightlines for the first(Srianand et al., 2008b) and second times (Noterdaeme et al.,2009a). We present here the third detection of CO, at z = .
69 to-wards SDSS J123714.60 + z em = .
78, hereafter calledJ 1237 +
2. Observations
We are conducting an observing campaign with X-shootermounted on the Cassegrain focus of the VLT Unit 2-Kueyen tele-scope to study the molecular content of our complete sample ofC absorbers. As a test case for the sensitivity of X-shooter inthe blue, we observed J 1237 + g = .
2) twice in servicemode on February 24 (airmass 1.2; seeing 1.4 ′′ ) and March 3,2010 (airmass 1.3; seeing 1.2 ′′ ), using a slit width of 1 ′′ in theUVB arm. Each observation run consisted in 1 h exposure takenin staring mode (see Table 1). This yields the nominal resolu-tion power of R = ∼
500 nm. Data were reduced using version0.9.5 of the preliminary ESO X-shooter pipeline (Goldoni et al.,2006) and the appropriate calibration data. The two individualspectra were then combined weighting each pixel by the inverseof the error variance. A portion of the X-shooter spectrum fea-turing CO is shown on Fig. 1, where several electronic bandsof CO are clearly detected. These bands are resolved into in-dividual rotational levels in the UVES spectrum obtained with8.5 h of exposure time (inset figures). From this, it is apparentthat X-shooter is the most e ffi cient instrument to survey a com-plete sample of candidates down to quasar magnitudes as faint as r ∼ .
5, whereas using UVES would be excessively time con-suming. Then, but only in the case of detection, higher spectralresolution is needed to make a detailed analysis of the physicalstate of the gas as done in the following.
The quasar J 1237 + × +
564 and 2 × +
2. Noterdaeme et al.: A translucent interstellar cloud at z = . Table 1.
Journal of observations
Instrument Date Setting Exposure time Resolving power a SNR b UVES 27-03-2009 390 +
564 2 × +
564 2 × +
775 2 × (cid:27) Notes. ( a ) ’B’, ’L’ and ’U’ stand for respectively blue, lower red and upper red CCD. ( b ) per pixel. N o r m a li s e d f l ux X−shooterCO AX(4−0) CO AX(3−0) CO AX(2−0)
UVES
UVES
UVES
Fig. 1.
Portion of X-shooter spectrum in the UVB. The inset fig-ures show 5 Å-wide portions of the UVES spectrum around theposition of the detected electronic bands of CO.to cover the wavelength range 3300-9600 Å with small gaps at4517-4621, 5597-5677 and 7764-7809 Å. The CCD pixels werebinned 2 × ′′ , yielding a re-solving power of ∼
50 000 under seeing conditions of 0.9-1 ′′ .Individual science spectra were reduced using the ESO UVESpipeline, which performs accurate sky subtraction while remov-ing cosmic ray impacts at the same time. The spectra were thencombined using a dedicated IDL routine by weighting each pixelby the inverse of the error variance in that pixel and clippingresidual cosmic rays impacts that remained after the cleaning of2D spectra.
3. Analysis
The system at z = .
69 towards J 1237 + , O , C , Mg , Cl andS ), singly-ionised (Fe + , Si + , Zn + , Ni + , S + , C + ), and molecularspecies (two isotopomers of molecular hydrogen: H and HD; aswell as carbon monoxide: CO).We analysed the UVES spectrum using standard Voigt pro-file fitting techniques. The fits were performed through χ -minimisation using the code FITLYMAN (Fontana & Ballester,1995) which is available as a context of the ESO-MIDASdata analysis software. The spectrum was normalised in thewavelength ranges of interest by fitting spline functions toregions free from absorption lines. Atomic data were takenfrom Morton (2003) for metal lines, unless otherwise spec-ified. Wavelengths and oscillator strengths were taken fromMorton & Noreau (1994) and Eidelsberg & Rostas (2003) forCO and from Abgrall & Roue ff (2006) for HD. Updated wave-lengths of H Lyman and Werner bands were taken fromBailly et al. (2010), with oscillator strengths from the Meudongroup , based on calculations described in Abgrall et al. (1994). http://amrel.obspm.fr/molat/ Photospheric solar abundances are taken from Asplund et al.(2009).
From the damped Lyman- α absorption line (see Fig. 2), wemeasure the total column density of atomic hydrogen to belog N (H )(cm − ) = . ± .
15, which is in agreement withthe value measured automatically by Noterdaeme et al. (2009b)from the low resolution SDSS spectrum (20 . ± . i profile is well constrained by the Lyman- β and Lyman- γ absorption lines. The large Doppler parameter( b ∼
100 km s − ) required to fit the Lyman- β and Lyman- γ lines is a consequence of the presence of multiple componentsas testified by the clumpy profile of the O i λ ∼
350 km s − (see Fig. 2). We recall that O closely follows H because of favourable charge-exchange re-action. Unfortunately, because of strong saturation and blendinge ff ects, it is not possible to derive column densities in individualcomponents and only the total H column density along the lineof sight is accessible.In Fig. 2 and subsequent figures and tables, the zero of thevelocity scale is taken at the position of the CO component( z abs = components at z abs = ∆ v = − − − − , see Sect. 3.3) are indicated byshort vertical marks. Interestingly, the centroid of the atomichydrogen absorption profile (vertical dotted line in Fig. 2 at z abs = + ±
10 km s − relative tothe CO absorption feature. This, the clumpy O i profile, and thelarge value of the b -parameter of H i lines, all indicate that asignificant fraction of the atomic gas is not associated with themolecular gas. We will discuss this further down in more details. Absorption lines from detected low ionisation species are spreadover about 350 km s − around the strongest component, which isalso the component where CO and HD are detected. We usednon-saturated transitions to derive accurate column densities forFe + , Ni + , S + and Zn + . Lines from other species (C + , O , N ) areheavily saturated, preventing us to derive any meaningful valueof the corresponding column densities. It is however possibleto perform an accurate measurement of the Si + column densityfrom the simultaneous use of Si ii λ σ detection level), and Si ii λ ii λ
3. Noterdaeme et al.: A translucent interstellar cloud at z = . λ λ γ HI Ly− γ β HI Ly− β −1 )0.00.51.0 HI Ly− α HI Ly− α Fig. 2.
Measurement of the total column density of neutralatomic hydrogen at z = .
69 towards J 1237 + -bearing components, the reddest of which also fea-tures CO and HD absorptions. The origin of the velocity scale,for this figure and all following ones, is defined at the positionof the CO-bearing component at z abs = ii λ ii λ ii λ σ upper-limits on the columndensities of Ni + and Zn + for the undetected components are pro-vided in the table. Finally, we measure log N (Cr + ) < . σ from the non-detection of Cr ii λ ff ect of ionisation on the overall abundances should benegligible. Indeed, even in the general population of absorbers,the ionisation correction is only about 0.1 dex for N (H ) = cm − (P´eroux et al., 2007). The metallicity is super-solarwith [Zn / H] = + .
34 and [S / H] = + .
15. Other species aredepleted ([Fe / Zn] = − / Zn] = − As the ionisation potential of neutral carbon (C ) is similar to theenergy of the photons that destroy H , C i is usually a good tracerof the presence of H (Srianand et al., 2005). The expected po-sitions of several C i lines usually fall out of the Lyman- α for- Table 3.
Summary of overall gas-phase abundances
Species log N (cm − ) mean abundance a H ± . . . H + . − . h f H2 i = . ± / H = -5.92Zn + ± / H] = + ± b S + ± / H] = + ± b Fe + ± / H] = -1.05 ± b Si + ± / H] = -0.48 ± b Ni + ± / H] = -0.86 ± b Notes. ( a ) Abundances are given considering the total neutral hydrogencolumn density N (H) = N (H ) + N (H ). ( b ) Relative to solar abundances(Asplund et al., 2009). est. We therefore initiated a program to search for moleculesalong QSOs selected upon the presence of C i , as seen in thelow resolution SDSS spectra. Because of this selection, it is notsurprising to detect strong C i lines in the UVES spectrum ofJ 1237 + i absorption lines is complex andresults from the blending of absorption lines from di ff erent com-ponents seen in di ff erent excitation levels (ground state: P , firstexcited level: P and second excited level: P ). Nevertheless,the high signal-to-noise and high spectral resolution allow us toclearly identify eight components. Most of them are also de-tected in the first excited level, while only the strongest twoare detected in the second excited level. The fit to C i lines isshown on Fig. 4, with the corresponding parameters given inTable 4. We considered all optically thin absorption lines but didnot include in the fit weak absorption lines in the region aroundC i λ column densitiescomes from the uncertainties on the oscillator strengths. As inprevious works from our group (e.g. Noterdaeme et al., 2007a,b;Srianand et al., 2008b) and others in the field (Jorgenson et al.,2009), we used f -values from Morton (2003). Using the oscilla-tor strengths from Jenkins & Tripp (2001) results in 2 to 3 timeslower column densities.The relative populations of the fine-structure levels of neu-tral carbon depend on the gas pressure. Since the kinetic tem-perature of the gas can be derived from the the relative pop-ulations of the low rotational levels of H (see Sect. 3.3),it is possible to measure the volumic density of the gas.From figure 2 of Silva & Viegas 2002 (see also Srianand et al.,2000) and taking into account excitation by collisions andby the Cosmic Microwave Background radiation, we can seethat the measured ratios log N (C ,J = / N (C ,J = = − N (C ,J = / N (C ,J = = − n H ∼ − for T ∼
110 K (seeSect. 3.3). Other components have similar fine-structure ratios,which indicate similar thermal pressure. The kinetic temperatureis probably larger in all other components (which is verified atleast for the two other H -bearing components, see Table 6), im-plying smaller densities (e.g. n H ∼ − in the componentat v = -127 km s − ). The first ionisation potential of sulphur being of 10.36 eV,this element is hence usually observed in its first ionisedstate (S + ) in Damped Lyman- α systems. In turn, sulphur is
4. Noterdaeme et al.: A translucent interstellar cloud at z = . λ λ λ λ λ λ λ λ λ λ −1 )0.80.91.0 NiII λ −1 )0.00.51.0 SiII λ Fig. 3.
Fit to metal lines. The origin of the velocity scale is set at the redshift of the CO absorption ( z = . components are indicated by short tick marks. The absorption seen at v = +
55 km s − in the Zn ii λ i λ v = −
110 km s − on the Si ii λ i λ ii lines are a ff ectedby blends. Finally Si ii λ ) inside molecularclouds, where the surrounding UV field has been strongly at-tenuated. To date, S has been detected only in QSO ab-sorbers where CO absorption is seen as well: in the sys-tems at z = + + z = + + (Srianand et al. 2005, albeit with generallymodest molecular fractions, Noterdaeme et al. 2008a), S i linesmight well indicate the presence of CO. This is however of littlepractical use for pre-selecting CO-bearing DLAs from the lowresolution SDSS spectra, since S i and CO lines are located inthe same spectral region and have similar strengths.We used all detected S i lines to constrain the column densityand b parameter. Two components are needed to properly fit thedata, with resulting χ ν ≃
1. The b -value obtained is less than1 km s − for the main component, i.e., well below the spectralresolution ( ∼ − ). However, b is well constrained, thanks tothe relatively large range spanned by the oscillator strength val- Table 5. S column densities z abs ∆ v a (km s − ) log N (S ) (cm − ) b (km s − )2.68953 -3 12.29 ± ± + ± ± Notes. ( a ) Relative to z abs = . ues. However, the fit is sensitive to the exact value of the spectralresolution. Therefore, to add confidence to the b and N measure-ments, we built the curve of growth for the detected S i lines,which does not depend on the spectral resolution (see Fig. 6).The error on the equivalent width measurements are conserva-tive and take into account uncertainties in the continuum place-ment. From this figure, we confirm the small Doppler parameter.The measured column density nicely matches the sum of the in-dividual column densities in the two components derived fromthe Voigt-profile fitting.
5. Noterdaeme et al.: A translucent interstellar cloud at z = . Table 2.
Column densities of metal species z abs ∆ v a (km s − ) b (km s − ) log N b (cm − )Fe + Ni + S + Zn + Si + Mg ± ± ± ± ± ± < ± ± ± ± ± ± < ± ± < ± ± ≤ < ± ± ± ± ± ≤ < ± ± ± ± ± ± < + ± ± < ± < ± < + ± ± ± ± ± ± ± +
55 3.5 ± ± < ≤ < ± < +
71 14.8 ± ± ± ≤ < ± < +
105 9.3 ± ± < ≤ ± ± < +
115 3.7 ± ± ± ≤ ± ± < +
136 3.4 ± ± < ≤ ± ± < +
188 19.5 ± ± < < < ± < +
214 6.5 ± ± < ± < ± < Notes. ( a ) In all figures and tables, the velocity is given with respect to the redshift of the CO component at z = . ( b ) Upper-limits due toblends or saturated lines are indicated by ’ ≤ ’, while 3 σ upper-limits from non-detections are indicated by ’ < ’. Table 4.
Column densities of neutral carbon in fine-structure levels z abs ∆ v a b log N (C ) (cm − )(km s − ) (km s − ) P P P total2.68802 -126 1.2 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± + ± ± ± ± ± +
114 3.6 ± ± ± ± Notes. ( a ) Relative to z abs = . Chlorine, with an ionisation potential of 12.97 eV is a uniquespecies among those that can be photoionised by photons withenergy h ν < . + whenhydrogen is mostly in the atomic form. However, neutral chlo-rine (Cl ) results from rapid exothermic ion-molecule reactionbetween singly-ionised chlorine (Cl + ) and H when H is opti-cally thick (Jura, 1974). Therefore, the presence of neutral chlo-rine in the ISM is expected to be a good indicator of the presenceof molecular gas.Cl is clearly detected at z abs = component. We measure log N (Cl ) = ± i λ b = . ± . − (see Fig 7). From the non-detection of Cl ii λ N (Cl + ) < . σ confidence level, whichtranslates to f Cl ≡ N (Cl ) / ( N (Cl ) + N (Cl + )) > .
3. As thefraction of chlorine in neutral form is expected to follow ap-proximately that of hydrogen in molecular form (Jura & York1978, see also Sonnentrucker et al. 2002), the lower limit on f Cl indicates that hydrogen could be mostly molecular at the placewhere we detect Cl . We indeed show in the next Section thatthe molecular fraction is particularly high in this component. Molecular hydrogen is detected in three components at z abs = ∼
125 km s − . The strongest component at z abs = Σ u + ← X Σ g + ) as well as some Werner bands(C Π u + ← X Σ g + ), which allows for an accurate measurementof the H column densities in each component and in di ff erentrotational levels. A portion of the UVES spectrum covering theH Lyman (1-0) band is shown on Fig. 8, while the full veloc-ity plots for di ff erent rotational levels are shown on Figs. 16 to21. The measured column densities and corresponding excitationtemperatures are given in Table 6). In the following, we refer tothese components as J = = b ∼ − ) is likely to be dominated by turbulent motions.Component ff erent Doppler pa-rameters. This behaviour has already been observed in theGalactic ISM (e.g. Jenkins & Peimbert, 1997; Lacour et al.,2005b) but also in high redshift Damped Lyman- α systems(Noterdaeme et al., 2007a). Doppler parameters can be mea-sured accurately even when significantly smaller than the spec-tral resolution thanks to the presence of numerous transitionswith di ff erent oscillator strengths. However, the measurementof b in the first rotational level (J =
0) remains di ffi cult due tothe small number of unblended absorption lines (see Table 6).However, there are enough transitions from the J =
6. Noterdaeme et al.: A translucent interstellar cloud at z = . CI λλ CI λ CI λ CI λ CI λ Observed wavelength (Å) N o r m a li s e d f l ux Fig. 4.
Portions of the J 1237 + i absorption lines. The short vertical marks in each panel representthe positions of absorption lines from the ground state ( P ),the first excited level ( P ) and the second excited level ( P )from top to bottom, respectively. The component where CO isdetected (at z abs = λ λ λ λ λ λ −1 )0.500.751.00 SI λ −1 )0.500.751.00 SI λ Fig. 5.
Fit to S i absorption lines. Results of the best model fitusing two components is overplotted ( χ ν = . −7.0 −6.5 −6.0 −5.5 −5.0log f λ (cm −1 )−6.0−5.8−5.6−5.4−5.2−5.0 l og W / λ −1 ) 12.612.813.013.213.413.613.814.0 l og N ( c m − ) σ σ Fig. 6.
Curve of growth analysis of S i absorption lines. Left:Curve of growth. Right: confidence interval. Minimum χ ν (0.91)is reached for log N (S ) = . b = .
91 km s − . −100 −50 0 50 100Relative velocity (km s −1 )0.500.751.00 ClI λ Fig. 7. Cl i λ z abs = b ∼ − . This is consistent with thermal excitation with atemperature of T k ∼
120 K, which is similar to what is measuredfrom T .The reddest component ( =
3, allowing for accu-rate measurement of the column densities. Non-saturated linesfor J = column density, but because it alsocontains deuterated molecular hydrogen, carbon monoxide aswell as neutral sulphur and neutral chlorine, all of which havebeen very rarely detected at high redshift. Since the column den-sity of H is large, the J = = N (J = / N (J =
0) ratio ismaintained at the Boltzmann equilibrium value. This means thatthe measurement of the kinetic temperature from T is robust.We measure T = T kin ∼
110 K, which is similar to the temper-ature in the local interstellar medium ( T kin ∼
80 K; Savage et al.,1977).We measure a total column density log N (H ) = . + . − . in the sub-DLA system with far the most important contribu-tion coming from the CO-bearing component (component h f H2 i = N (H ) / (2 N (H ) + N (H )) = . + . − . . However, the centre ofthe H i Lyman- α absorption line is clearly shifted from compo-nent − (see Fig. 2). This means that theamount of atomic hydrogen in the CO-bearing cloud is much
7. Noterdaeme et al.: A translucent interstellar cloud at z = . Table 6. H column densities and excitation temperatures component log N (H ,J) a b T − J z abs , v (km s − ) (cm − ) (km s − ) (K) z = ∆ v = -127 16.28 + . − . J = ± ± = ± + − J = ± + − J = ± + − z = ∆ v = -73 17.62 + . − . J = + . − . + . − . –J = ± ± ≥ = ± ± + − J = ± ± + − z = ∆ v = -2 19.20 + . − . J = ± ± = ± + − J = ± + − J = ± + − J = ± ± + − J = ± + − Notes. ( a ) The first line for each component gives the total H columndensity in that component. smaller than log N (H ) =
20 and the value given above shouldbe considered as a lower limit on the actual molecular fraction inthe CO-bearing cloud (i.e. f H2 > / absorption associated with Several Lyman-band HD lines from the first two rotational levelsare detected in the UVES spectrum (see Fig. 9). Unfortunately,J = = N (HD) / N (H ) = . × − . This is about10 times higher than what is measured in the Galactic ISM(Lacour et al., 2005a) for f = .
24. Since this ratio is knownto increase with the molecular fraction (Lacour et al., 2005a)HD / might provide a lower limit on D / H for f H2 <
1. If, asdiscussed previously, the actual molecular fraction in the HD-bearing cloud towards J 1237 + could be self-shielded and HD / ∼ D / H . Thevalue we obtain is then consistent with the D / H ratio mea-sured in the Galactic disc (Linsky et al., 2006). This correspondsto an astration factor of ∼ / H = ± × − ; Pettini et al. 2008, Ivanchik et al. 2010,see however Srianand et al. 2010) or derived from the baryondensity parameter (Steigman, 2007). Table 7.
HD column densities component log N (HD,J) (cm − ) b (km s − ) z abs = ∆ v = -1 km s − J = ± .
05 4.5 ± = ≤ .
60 ”
Note that all five high redshift HD detections to date yieldrelatively large D / H values despite significant metal enrichment: N (HD) / N (H ) = . × − , 3 . × − , 7 . × − and1 . × − towards respectively, J 1439 + +
082 (Ivanchik et al., 2010), J 2123-0500 andFJ 0812 +
32 (Tumlinson et al., 2010). Since deuterium is easilydestroyed as interstellar gas is cycled through stars, large deu-terium abundances are di ffi cult to reproduce with closed-boxmodels. However, these are well explained by models includ-ing infall of primordial gas (e.g. Prodanovi´c & Fields, 2008).If the velocity-metallicity correlation found by Ledoux et al.(2006a) is the consequence of an underlying mass-metallicityrelation, then we can expect that a high astration of deuteriumin high metallicity systems is roughly compensated by a stronginfall of primordial material onto massive galaxies. However,Tumlinson et al. (2010) noted that HD / ratios in high- z ab-sorption systems lie in a narrow range well above the value mea-sured in the Galaxy while these systems present a large diver-sity in terms of metallicities and molecular fractions. This puz-zling behaviour led them to conclude that it could be prematureto use the HD / ratio to derive Ω b , given our actual under-standing of interstellar chemistry. In addition, we note that inthe case of QSO absorbers, we only have access to the proper-ties of the gas (metallicities, molecular fractions) averaged overthe line of sight. These may not be representative of the ac-tual chemical abundances in the HD-bearing cloud. Indeed, onlythe total N (H ) can usually be measured and the metal compo-nents are blended into a smooth absorption profile. It is thereforenecessary to be careful and to study each system in detail (e.g.Balashev et al., submitted) to derive the local chemical and phys-ical conditions in the cloud. Absorptions from eight A Π ( ν ′ ) ← X Σ + ( ν =
0) bands of CO(from AX (0-0) to AX (7-0)), the C Σ ( ν ′ = ← X Σ + ( ν = d ∆ ( ν ′ = ← X Σ + ( ν =
0) inter-band sys-tem are detected at z abs = + ff erent ro-tational levels in the P and R branches. The resolving power( R ∼
50 000) and the signal-to-noise ratio (SNR ∼
28) of theUVES spectrum are high enough to individually measure thecolumn densities in rotational levels up to J =
3. In addition, theJ = CX (0-0). Weuse the AX and dX bands that fall outside of the Lyman- α forestto measure the column densities in rotational levels up to J = AX (3-0) band is a ff ected by a spike likely due to a cos-mic ray impact at the position of the R branch and this region istherefore not considered when fitting the profile. In addition, weuse the CX (0-0) band up to J =
4. This band is the strongest oneavailable but the corresponding rest wavelength (1088 Å) makesit redshifted in the Ly- α forest. Fortunately, only the R branch isblended whereas the P branch is free from any blend. Moreover,
8. Noterdaeme et al.: A translucent interstellar cloud at z = . R R P R P R R R P R P R R R P P R P R H L(1−0) P P P R P R Fig. 8.
Portion of the UVES spectrum of J 1237 + . The labels indicate the branches (’R’,’P’ for ∆ J = − , +
1, respectively) and the rotational levels of the lower states. Absorptions from di ff erent components are indicatedusing di ff erent label colours ( −1 )0.00.51.0 HD L7R00.00.51.0 −200 −150 −100 −50 0 50 100Relative velocity (km s −1 )0.00.51.0 HD L8R00.00.51.0 Fig. 9.
Fit to HD J = . The velocity of theH detected components are indicated by vertical dotted lines.the CX rotational levels are well separated at the UVES spectralresolution.The results of the fit are presented in Table 8. Two errorsare quoted in this table for the column densities: the first oneis the rms error on the fit, while the second one reflects the un-certainties resulting from the continuum placement. The latteruncertainties were estimated by changing the normalisation byplus or minus 0.5 σ (i.e. about ±
2% for SNR =
28) around thebest continuum fit. The total CO column density we derive is log N (CO) = ± ratio of N (CO) / N (H ) = − . This is typical of what is seen in translu-cent clouds (see Burgh et al., 2010). Table 8.
CO column densities and excitation temperatures component log N (CO,J) a b (km s − ) T − J b (K) z abs = . ± c ± = ± ± = ± ± + . − . J = ± ± + . − . J = ± ± + . − . J = ± ± + . − . Notes. ( a ) Quoted errors on column densities are respectively errorsfrom fitting the Voigt profiles and errors due to continuum placement.The latter were estimated by varying the continuum by ± . σ . ( b ) Theerrors on T − J represent the extremum values for the di ff erent sets ofcontinuum. ( c ) Total CO column density.
In Fig. 11, we show the excitation diagram of CO. It isclear that the population of the first three rotational levels canbe reproduced with a single excitation temperature. We mea-sure this excitation temperature by performing a linear fit oflog N (CO,J) / g J vs the energy of the levels ( E ). The fit and1 σ range on Fig. 11 corresponds to the best fit continuum. Inorder to estimate the e ff ect of the continuum placement, we re-peat the linear fit for each set of continua and take the extremaas representative of the range of possible values for T ex (CO).This gives T ex = . + . − . . Note that the e ff ect of the continuumplacement is mainly a change in the total CO column density,while little change on the slope of the linear fit (i.e. the exci-tation temperature). The CO excitation temperature is well be-low the kinetic temperature of the gas. This means that the gas
9. Noterdaeme et al.: A translucent interstellar cloud at z = . R R R R P P P P R CO CX(0−0) R R R R Q Q Q P P CO AX(0−0) R R R R Q Q Q P P CO AX(1−0) R R R R Q Q Q P P CO AX(2−0) R R R R Q Q Q P P CO AX(3−0) R R R R Q Q Q P P CO AX(4−0) R R R R Q Q Q P P CO AX(5−0) R R R R Q Q Q P P CO AX(6−0) R R R R Q Q Q P P CO AX(7−0)
Observed wavelength (Å) N o r m a li s e d f l ux Fig. 10.
Fit to CO lines ( χ ν = . ff use molec-ular and translucent clouds (e.g. Warin et al., 1996). Indeed, thepopulation ratios of the neutral carbon fine-structure levels, in-dicate a volumic density of the order of n H ∼
50 cm − , wellbelow the critical density at which the collisional de-excitationrate of CO(J = n crit ∼ − ; Snow & McCall, 2006). Indeed, in terms ofdensity and molecular fraction, the CO-bearing system presentedhere is very similar to that presented in Srianand et al. (2008b)where we concluded that collisional excitation of CO is negligi-ble. It is important to note however, that the fine-structure levelsof C only give the average volumic density. The actual local vo-lumic density in the CO-bearing cloud could be higher. A smallshift ( ∼ − ) is measured between the strongest C i featureand the CO component. This may indicate that the two speciesare not completely co-spatial.From the radiative code RADEX (van der Tak et al., 2007),we expect the excitation temperature of CO to be about onedegree larger than the expected temperature of the CosmicMicrowave Background (CMB) radiation ( T CMB ( z = . = .
05 K) as soon as the collision partner (H , H and He) densityis larger than 50 cm − . This explains that the excitation temper-ature we measure is slightly higher than what is expected fromexcitation by the CMB radiation alone. (K)11.011.512.012.513.013.514.0 l og N ( C O , J ) / g J Fig. 11.
Excitation diagram of CO rotational levels. Errors onthe column densities from fitting the lines are represented by thesmall red error bars while the long black error bars take into ac-count the uncertainty in the continuum placement. The plain linerepresents the linear regression fit using J = = σ errors are represented bydashed lines.In Fig. 12, we compare the excitation temperature of CO athigh redshift with that in the local Universe. In the local ISM,the temperature is seen a few degrees above T CMB at low COcolumn densities and rises for column densities above N (CO) = × cm − . This is due to the increased importance of pho-ton trapping at larger column densities (Burgh et al., 2007). Thevalues observed at high redshift are significantly higher than thelocal ones, despite similar N (CO) and kinetic temperatures. Thisclearly means that the main physical di ff erence between highredshift and local lines of sight is the higher CMB temperatureat high redshift. This provides a strong positive test to the hotBig-Bang theory. Another consequence of Fig. 12 is that onlyCO-bearing systems with log N (CO) <
15 – for which there isno correlation between N (CO) and T ex (CO) – are good placeswhere to measure the evolution of T CMB with cosmic time.Interestingly, although the di ff erences are small and withinerrors, we measure a systematic trend, T > ∼ T > ∼ T > ∼ T ,regardless of the exact continuum placement. This indicates thatwhile CMB photons dominate the rotational excitation of CO,other mechanisms are at play. We fail to reproduce the increas-ing temperature with increasing rotational level with RADEX.However, such behaviour has already been noticed in the lo-cal ISM (Sonnentrucker et al., 2007; She ff er et al., 2008) andcould be explained by the selective self-shielding of low ro-tational lines for log N (CO) >
14 (Warin et al., 1996). The self-shielding of far-UV Rydberg bands of CO (those relevant to thephoto-destruction process) could be more e ff ective than previ-ously thought (She ff er et al., 2003). In addition, the presence ofH lines in the same spectral region can contribute to an e ff ec-tive shielding of CO lines. Finally, radiative pumping from COemission lines due to nearby dense molecular clouds could con-tribute to populate the higher rotational levels in the absorbingcloud (Wannier et al., 1997). If the increasing temperature withincreasing rotational level is physical, then T = . + . − . Kcould represent better the excitation by the CMB alone.
10. Noterdaeme et al.: A translucent interstellar cloud at z = .
13 14 15 16log N(CO)24681012 T e x ( C O ) ( K ) T CMB
Fig. 12.
Excitation temperature of CO as a function of the to-tal CO column density. Black error bars are measurements at z = z = . + z = . + T ex (6-16 K) could be determined for the system at z = .
64 towardsSDSS J160457.50 + N (CO) and T ex (CO).
4. The nature of the absorbing cloud
In the previous sections, we have derived physical proper-ties of the gas associated to the molecular absorptions seen at z abs ∼ + , we derived that the molecular frac-tion, 2 N (H ) / (2 N (H ) + N (H )), in the CO component is largerthan 1 / Z (Zn,S) = + + ground-state fine structure levels,we found that the particle density is of the order of ∼
50 cm − .The analysis of the H rotational levels yields a kinetic tempera-ture of ∼
100 K and CO is mainly excited by radiative interactionwith the CMBR.We can have an indication of the electronic density in thecloud thanks to the S / S + ratio. Assuming that the mean ratiocan be derived using the column densities in the strongest com-ponents we measure: log N (S ) / N (S + ) = − .
72. The electronicdensity, n e , is derived from the ionisation equilibrium betweenthe two species Γ n (S ) = α n e n (S + ) , (1)where Γ is the photoionisation rate of S and α the combina-tion rate of S + . Taking the ratio in di ff use gas of the Galaxy( Γ /α ∼
95 cm − , P´equignot & Aldrovandi 1986; see also table 8of Welty et al. 1999b) we derive n e ∼ Γ /α /
95) cm − . Notethat this is an average value in the strongest metal component.Since S and S + are not co-spatial –as indicated by the di ff erentDoppler-parameters– the electron density derived here should beconsidered as an upper limit in the molecular gas. −0.2 0.0 0.2 0.4 0.6 0.8 1.0g−r020406080 N u m b e r Fig. 13.
Distribution of SDSS photometric g − r values for a sam-ple of 650 non-BAL QSOs at 2 . < z < . + + column density as C ii λ N (C + ) from the sulphur electronic densitymeasurement. We find N (C + ) N (S + ) = Γ (C ) /α (C + ) Γ (S ) /α (S + ) N (C ) N (S ) (2)Using Γ (C ) /α (C + ) =
32 cm − yields log N (C + ) / N (S + ) = . Fig. 13 shows the distribution of g − r colours for 650 non-BALquasars with redshifts similar to J 1237 + ∆ z = . g − r value of the distribution is 0.16 with a standard de-viation of 0.11. This dispersion reflects the color variation fromone QSO to the other and is not due uncertainties in the SDSSphotometric measurements (which are better than 0.03 mag).This implies that the colour excess of J 1237 + g − r = . σ level. This significance increases to4.9 σ if we consider a Gaussian fit to the distribution. Indeed,there is a tail towards large colour excesses which shows theexistence of a population of reddened quasars. J 1237 + g − r > . + = ± + α emission.In order to further estimate the probability that the abovereddening might be due to a peculiar intrinsic QSO shape,we used a technique described in (Srianand et al., 2008a;Noterdaeme et al., 2009a): We repeated the spectral slope fitting
11. Noterdaeme et al.: A translucent interstellar cloud at z = . assuming an absorber at z = .
69 for a control sample of 82 non-BAL QSOs with similar emission redshifts (2 . < z em < . i -band S / N ratios larger than 5. We find that thecontinuum slope of J 1237 + + z abs = . = = = = ± = ± > ff er fromflux overestimation which can easily explain the discrepancybetween predicted and measured H-band magnitudes. In addi-tion, the presence of the H- β emission line in the H-band in-creases the uncertainty of our estimate. This together with thefact that the 2MASS and SDSS observations were taken fiveyears apart, makes our predicted magnitudes consistent with the2MASS data. Unfortunately, there are no SDSS measurementsat di ff erent epochs for this object to monitor any variation in theQSO flux.The measured reddening, although significative, ismarginally higher to what is seen in the general popula-tion of DLAs (E(B-V) ∼ .
04, Ellison et al. 2005). Interestingly,the integrated extinction-to-gas ratio measured towardsJ 1237 + / N (H ) = × − mag cm is 20 timeshigher that the average value for the SMC (7.5 × − mag cm ;Gordon et al., 2003) and about 50 times higher than the meanvalue measured in high redshift DLAs (2-4 × − mag cm ;Vladilo et al., 2008). This means that would the H columndensity have been higher, the extinction induced would havebeen so large that the QSO would have been missed by theSDSS target selection . Note that the moderate extinction A V = .
14 in the rest-frame of the absorber, already produces anextinction of nearly 1 mag in the g -band. This supports furtherthe possibility that current surveys can miss a large number ofcold clouds (Noterdaeme et al., 2009a). If, as discussed before,a large fraction of the atomic hydrogen is not associated withthe molecular component, then the extinction-to-gas ratio inthe molecular component is even higher. The line-of-sight toJ 1237 + / Zn] ∼ -2.0,[Ni / Zn] ∼ -1.6, [Si / Zn] ∼ -1.7, [Cr / Zn] < − . < A V <
5. However, as notedby Rachford et al. (2002), even translucent sightlines can resultfrom the concatenation of multiple di ff use clouds along the line-of-sight. This kind of scenario has already been advocated toexplain the bimodal distribution in the log N (H ) distributionfor a given column density of dust (Noterdaeme et al., 2008a).Therefore extinction may not be the best parameter to definetranslucent clouds. Rachford et al. (2002) and Snow & McCall(2006) have proposed that the definition of translucent cloudsshould be based on the local properties of the gas rather than For log N (H ) = .
65 and same extinction-to-gas ratio, the pre-dicted magnitude already reaches the i = . F l ux ( − e r g s − c m − Å − ) QSO compositereddened QSO compositeQ1237+0647 SDSS spectrum
Fig. 14.
SDSS spectrum of J 1237 + = A V = In Fig. 15, we plot the ratios N (CO) / N (H ) and N (CO) / N (C )versus the hydrogen molecular fraction, f H2 , for the two systemstowards J 1237 + + / H > − and CO / C > f H2 > + N (CO) / N (H ) = − for h f H2 i = . We therefore conclude that the cloud in front ofJ 1237 +
5. Conclusion
From our VLT survey for H in DLAs, it appears that neutralcarbon is generally observed in the same components that fea-ture H (Srianand et al., 2005; Noterdaeme et al., 2008a). Wetherefore selected the rare SDSS lines of sight in which C i ab-sorptions are present. From UVES follow-up observations, wehave detected strong absorptions from H , HD and CO alongSDSS J123714.60 + column density is small, log N (H ) = ± / H] = + / H] =+ components spanning ∼
125 km s − , the strongest of which, Note that we use here log N (C , J = , , = . ± .
10 derivedusing f -values from Jenkins & Tripp (2001) to enable comparison withthe work by Burgh et al. (2010).12. Noterdaeme et al.: A translucent interstellar cloud at z = . −8 −7 −6 −5 −4 N ( C O ) / N ( H ) TranslucentDiffuse H2 N ( C O ) / N ( C ) TranslucentDiffuse
Fig. 15. N (CO) / N (H ) and N (CO) / N (C ) versus the molecularfraction, f H2 . The red filled circle with error bars is our mea-surement at z = . + z = . + ff use and translucent clouds, whilethe two vertical dotted lines mark the transition range betweenthese two regimes. Note that the molecular fraction we indicatestoward J 1237 + N (C ) measured with the f -values from Jenkins & Tripp (2001).with log N (H ) = i absorption. This means that the molecular fraction in thiscomponent is larger than the mean molecular fraction h f H2 i = / rotational levels, wemeasure the kinetic temperature of the gas to be around 100 K inthe strongest component, where HD and CO are also detected.The detection of S and C implies that the gas is shielded fromthe surrounding far-UV radiation field. The relative populations of the C fine structure levels yieldsan estimate of the average hydrogen density in the main com-ponent of about 50-60 cm − . At such densities, collisions arenot frequent enough to dominate the rotational excitation ofCO molecules and radiative processes are likely to determinethe CO rotational populations. The excitation temperature wemeasure ( T ex (CO) = T ex (CO) ∼ . + T CMB ( z = . = − ) are observedbetween the di ff erent neutral and molecular species in the maincomponent. For H , the shift might appear more important butis rather due to the H component being a blend of severalsub-components. Neutral chlorine is likely a better indicator ofthe strongest molecular component (Sonnentrucker et al., 2006).Srianand et al. (2010) recently observed similar small velocityshifts between H and 21-cm absorption in a high redshift DLAalso testifying the presence of inhomogeneities of the ISM onvery small scales. In turn, since the distribution of H and O are expected to closely follow each other, the O i λ ∼
400 km s − velocity range over which metals are observed.This explains the large velocity shift between the centroid of H i and molecular lines.All this reinforces the view that the ISM is patchy, withsmall and dense molecular cloudlets (probably with an onion-like structure) immersed in a warmer di ff use atomic medium (seee.g. Petitjean et al., 1992). We can derive an upper limit on thesize of the molecular-rich region along the line-of-sight by con-sidering that all H is associated with the main velocity compo-nent of density n H =
50 cm − . The corresponding characteristiclength is l = N (H ) / n H ≃ . l > .
05 pc. The molecular region of the systemhas therefore a very small size and hence small cross-section.It is therefore not surprising that detections of translucent cloudswere elusive till now. Studying the frequency of CO absorberswould give an idea of the filling factor of the molecular-richgas, but requires larger statistics. Small physical extents couldyield partial covering of the background source by the cloud.This may happen in particular if some absorption lines are red-shifted on top of emission lines from the extended QSO broadline region, as seen in the case of Q 1232 +
082 (Ivanchik et al.2010; Balashev et al., submitted).Interestingly, the conclusion that the ISM at high redshiftis made of small cold cloudlets immersed in warmer di ff usemedium has been reached by Gupta et al. (2009) while consid-ering the distribution of cold gas detected in 21-cm absorption.Note that J 1237 + + (e.g. Dalgarno et al., 1974; Neufeld & Wolfire,2009). The DLA system toward J 1237 + , CH, OH and
13. Noterdaeme et al.: A translucent interstellar cloud at z = . study astrochemistry in the interstellar medium of high redshiftgalaxies. Given the expected attenuation of the quasar by CO-bearing clouds, X-shooter, with its high throughput and mediumresolution, is the best instrument to survey carefully selectedDLA / sub-DLA candidates for CO absorptions and build a sam-ple of molecular-rich clouds at high redshift, which may then bestudied in details in the optical and sub-millimeter ranges withfuture facilities like ELTs and ALMA. Acknowledgements.
PN is supported by a CONICYT / CNRS fellowship andgratefully acknowledges the European Southern Observatory for hospitality dur-ing part of the time this work was done. SL is supported by FONDECYT grantN o References
Abgrall, H. & Roue ff , E. 2006, A&A, 445, 361Abgrall, H., Roue ff , E., Launay, F., & Roncin, J.-Y. 1994, Canadian Journal ofPhysics, 72, 856Asplund, M., Grevesse, N., Sauval, A. J., & Scott, P. 2009, ARA&A, 47, 481Bailly, D., Salumbides, E. J., Vervloet, M., & Ubachs, W. 2010, MolecularPhysics, inpressBalashev, S. A., Petitjean, P., Ivanchik, A. V., et al. 2010, submitted to MNRASBurgh, E. B., France, K., & Jenkins, E. B. 2010, ApJ, 708, 334Burgh, E. B., France, K., & McCandliss, S. R. 2007, ApJ, 658, 446Cui, J., Bechtold, J., Ge, J., & Meyer, D. M. 2005, ApJ, 633, 649Dalgarno, A., de Jong, T., Oppenheimer, M., & Black, J. H. 1974, ApJ, 192, L37Dekker, H., D’Odorico, S., Kaufer, A., Delabre, B., & Kotzlowski, H. 2000, inProc. SPIE Vol. 4008, p. 534-545, Optical and IR Telescope Instrumentationand Detectors, Masanori Iye; Alan F. Moorwood; Eds., 534–545Eidelsberg, M. & Rostas, F. 2003, ApJS, 145, 89Ellison, S. L., Hall, P. B., & Lira, P. 2005, AJ, 130, 1345Fontana, A. & Ballester, P. 1995, The Messenger, 80, 37Fynbo, J. P. U., Jakobsson, P., Prochaska, J. X., et al. 2009, ApJS, 185, 526Fynbo, J. P. U., Starling, R. L. C., Ledoux, C., et al. 2006, A&A, 451, L47Goldoni, P., Royer, F., Franc¸ois, P., et al. 2006, in Society of Photo-OpticalInstrumentation Engineers (SPIE) Conference Series, Vol. 6269, Society ofPhoto-Optical Instrumentation Engineers (SPIE) Conference SeriesGordon, K. D., Clayton, G. C., Misselt, K. A., Landolt, A. U., & Wol ff , M. J.2003, ApJ, 594, 279Gupta, N., Srianand, R., Petitjean, P., Noterdaeme, P., & Saikia, D. J. 2009,MNRAS, 398, 201Hirashita, H. & Ferrara, A. 2005, MNRAS, 356, 1529Ivanchik, A. V., Petitjean, P., Balashev, S. A., et al. 2010, MNRAS, 297Jenkins, E. B. & Peimbert, A. 1997, ApJ, 477, 265Jenkins, E. B. & Tripp, T. M. 2001, ApJS, 137, 297Jorgenson, R. A., Wolfe, A. M., Prochaska, J. X., & Carswell, R. F. 2009, ApJ,704, 247Jura, M. 1974, ApJ, 190, L33Jura, M. & York, D. G. 1978, ApJ, 219, 861Kanekar, N. & Chengalur, J. N. 2003, A&A, 399, 857Lacour, S., Andr´e, M. K., Sonnentrucker, P., et al. 2005a, A&A, 430, 967Lacour, S., Ziskin, V., H´ebrard, G., et al. 2005b, ApJ, 627, 251Ledoux, C., Petitjean, P., Fynbo, J. P. U., Møller, P., & Srianand, R. 2006a, A&A,457, 71Ledoux, C., Petitjean, P., & Srianand, R. 2003, MNRAS, 346, 209Ledoux, C., Petitjean, P., & Srianand, R. 2006b, ApJ, 640, L25Ledoux, C., Vreeswijk, P. M., Smette, A., et al. 2009, A&A, 506, 661Linsky, J. L., Draine, B. T., Moos, H. W., et al. 2006, ApJ, 647, 1106Markwardt, C. B. 2009, ArXiv e-prints 0902.2850Morton, D. C. 2003, ApJS, 149, 205Morton, D. C. & Noreau, L. 1994, ApJS, 95, 301Neufeld, D. A. & Wolfire, M. G. 2009, ApJ, 706, 1594Noterdaeme, P., Ledoux, C., Petitjean, P., et al. 2007a, A&A, 474, 393Noterdaeme, P., Ledoux, C., Petitjean, P., & Srianand, R. 2008a, A&A, 481, 327Noterdaeme, P., Ledoux, C., Srianand, R., Petitjean, P., & Lopez, S. 2009a,A&A, 503, 765Noterdaeme, P., Petitjean, P., Ledoux, C., & Srianand, R. 2009b, A&A, 505,1087Noterdaeme, P., Petitjean, P., Ledoux, C., Srianand, R., & Ivanchik, A. 2008b,A&A, 491, 397Noterdaeme, P., Petitjean, P., Srianand, R., Ledoux, C., & Le Petit, F. 2007b,A&A, 469, 425 P´equignot, D. & Aldrovandi, S. M. V. 1986, A&A, 161, 169P´eroux, C., Dessauges-Zavadsky, M., D’Odorico, S., Kim, T.-S., & McMahon,R. G. 2007, MNRAS, 382, 177P´eroux, C., Dessauges-Zavadsky, M., Kim, T., McMahon, R. G., & D’Odorico,S. 2002, Ap&SS, 281, 543Petitjean, P., Bergeron, J., & Puget, J. L. 1992, A&A, 265, 375Petitjean, P., Ledoux, C., Noterdaeme, P., & Srianand, R. 2006, A&A, 456, L9Petitjean, P., Srianand, R., & Ledoux, C. 2000, A&A, 364, L26Petitjean, P., Srianand, R., & Ledoux, C. 2002, MNRAS, 332, 383Pettini, M., Zych, B. J., Murphy, M. T., Lewis, A., & Steidel, C. C. 2008,MNRAS, 391, 1499Prochaska, J. X., She ff er, Y., Perley, D. A., et al. 2009, ApJ, 691, L27Prodanovi´c, T. & Fields, B. D. 2008, Journal of Cosmology and Astro-ParticlePhysics, 9, 3Rachford, B. L., Snow, T. P., Tumlinson, J., et al. 2002, ApJ, 577, 221Reimers, D., Baade, R., Quast, R., & Levshakov, S. A. 2003, A&A, 410, 785Savage, B. D., Bohlin, R. C., Drake, J. F., & Budich, W. 1977, ApJ, 216, 291She ff er, Y., Federman, S. R., & Andersson, B. 2003, ApJ, 597, L29She ff er, Y., Rogers, M., Federman, S. R., et al. 2008, ApJ, 687, 1075Silva, A. I. & Viegas, S. M. 2002, MNRAS, 329, 135Snow, T. P. & McCall, B. J. 2006, ARA&A, 44, 367Sonnentrucker, P., Friedman, S. D., Welty, D. E., York, D. G., & Snow, T. P.2002, ApJ, 576, 241Sonnentrucker, P., Friedman, S. D., & York, D. G. 2006, ApJ, 650, L115Sonnentrucker, P., Welty, D. E., Thorburn, J. A., & York, D. G. 2007, ApJS, 168,58Srianand, R., Gupta, N., Petitjean, P., Noterdaeme, P., & Ledoux, C. 2010,MNRAS, 405, 1888Srianand, R., Gupta, N., Petitjean, P., Noterdaeme, P., & Saikia, D. J. 2008a,MNRAS, 391, L69Srianand, R., Noterdaeme, P., Ledoux, C., & Petitjean, P. 2008b, A&A, 482, L39Srianand, R., Petitjean, P., & Ledoux, C. 2000, Nature, 408, 931Srianand, R., Petitjean, P., Ledoux, C., Ferland, G., & Shaw, G. 2005, MNRAS,362, 549Steigman, G. 2007, Annual Review of Nuclear and Particle Science, 57, 463Tumlinson, J., Malec, A. L., Carswell, R. F., et al. 2010, ApJ, 718, L156Tumlinson, J., Prochaska, J. X., Chen, H.-W., Dessauges-Zavadsky, M., &Bloom, J. S. 2007, ApJ, 668, 667van der Tak, F. F. S., Black, J. H., Sch¨oier, F. L., Jansen, D. J., & van Dishoeck,E. F. 2007, A&A, 468, 627van Dishoeck, E. F. & Black, J. H. 1989, ApJ, 340, 273Vanden Berk, D. E., Richards, G. T., Bauer, A., et al. 2001, AJ, 122, 549Vladilo, G., Prochaska, J. X., & Wolfe, A. M. 2008, A&A, 478, 701Wannier, P., Penprase, B. E., & Andersson, B. 1997, ApJ, 487, L165Warin, S., Benayoun, J. J., & Viala, Y. P. 1996, A&A, 308, 535Welty, D. E., Frisch, P. C., Sonneborn, G., & York, D. G. 1999a, ApJ, 512, 636Welty, D. E., Hobbs, L. M., Lauroesch, J. T., et al. 1999b, ApJS, 124, 465Wolfe, A. M., Howk, J. C., Gawiser, E., Prochaska, J. X., & Lopez, S. 2004, ApJ,615, 625
14. Noterdaeme et al.: A translucent interstellar cloud at z = . L0R00.00.51.0 0.00.51.0 H L1R00.00.51.0 0.00.51.0 H L2R00.00.51.0 0.00.51.0 H L3R00.00.51.0 0.00.51.0 H L4R00.00.51.0 0.00.51.0 H L5R00.00.51.0 0.00.51.0 H L6R00.00.51.0 0.00.51.0 H L7R00.00.51.0 0.00.51.0 H L8R00.00.51.0 0.00.51.0 H L10R00.00.51.0 −200 −150 −100 −50 0 50 100Relative velocity (km s −1 )0.00.51.0 H W1R00.00.51.0 −200 −150 −100 −50 0 50 100Relative velocity (km s −1 )0.00.51.0 H W0R00.00.51.0
Fig. 16.
Fit to H (J =
0) lines. The blue profile is the contributionfrom HD. L0P10.00.51.0 0.00.51.0 H L0R10.00.51.0 0.00.51.0 H L1P10.00.51.0 0.00.51.0 H L1R10.00.51.0 0.00.51.0 H L2P10.00.51.0 0.00.51.0 H L2R10.00.51.0 0.00.51.0 H L3P10.00.51.0 0.00.51.0 H L3R10.00.51.0 0.00.51.0 H L4P10.00.51.0 0.00.51.0 H L4R10.00.51.0 0.00.51.0 H L5P10.00.51.0 0.00.51.0 H L5R10.00.51.0 0.00.51.0 H L6P10.00.51.0 0.00.51.0 H L6R10.00.51.0 0.00.51.0 H L7P10.00.51.0 0.00.51.0 H L7R10.00.51.0 0.00.51.0 H L8P10.00.51.0 0.00.51.0 H L8R10.00.51.0 0.00.51.0 H L9P10.00.51.0 0.00.51.0 H L9R10.00.51.0 0.00.51.0 H L10P10.00.51.0 0.00.51.0 H L10R10.00.51.0 0.00.51.0 H W0R10.00.51.0 0.00.51.0 H W0Q10.00.51.0 −200 −150 −100 −50 0 50 100Relative velocity (km s −1 )0.00.51.0 H W1R10.00.51.0 −200 −150 −100 −50 0 50 100Relative velocity (km s −1 )0.00.51.0 H W1Q10.00.51.0
Fig. 17.
Fit to H (J =
1) lines.
15. Noterdaeme et al.: A translucent interstellar cloud at z = . L0P20.00.51.0 0.00.51.0 H L0R20.00.51.0 0.00.51.0 H L1P20.00.51.0 0.00.51.0 H L1R20.00.51.0 0.00.51.0 H L2P20.00.51.0 0.00.51.0 H L2R20.00.51.0 0.00.51.0 H L3P20.00.51.0 0.00.51.0 H L3R20.00.51.0 0.00.51.0 H L4P20.00.51.0 0.00.51.0 H L4R20.00.51.0 0.00.51.0 H L5P20.00.51.0 0.00.51.0 H L5R20.00.51.0 0.00.51.0 H L6P20.00.51.0 0.00.51.0 H L6R20.00.51.0 0.00.51.0 H L7P20.00.51.0 0.00.51.0 H L7R20.00.51.0 0.00.51.0 H L8P20.00.51.0 0.00.51.0 H L8R20.00.51.0 0.00.51.0 H L9P20.00.51.0 0.00.51.0 H L9R20.00.51.0 0.00.51.0 H L10P20.00.51.0 0.00.51.0 H L10R20.00.51.0 0.00.51.0 H W0R20.00.51.0 0.00.51.0 H W0Q20.00.51.0 −200 −150 −100 −50 0 50 100Relative velocity (km s −1 )0.00.51.0 H W1R20.00.51.0 −200 −150 −100 −50 0 50 100Relative velocity (km s −1 )0.00.51.0 H W1Q20.00.51.0
Fig. 18.
Fit to H (J =
2) lines. L0P30.00.51.0 0.00.51.0 H L0R30.00.51.0 0.00.51.0 H L1P30.00.51.0 0.00.51.0 H L1R30.00.51.0 0.00.51.0 H L2P30.00.51.0 0.00.51.0 H L2R30.00.51.0 0.00.51.0 H L3P30.00.51.0 0.00.51.0 H L3R30.00.51.0 0.00.51.0 H L4P30.00.51.0 0.00.51.0 H L4R30.00.51.0 0.00.51.0 H L5P30.00.51.0 0.00.51.0 H L5R30.00.51.0 0.00.51.0 H L6P30.00.51.0 0.00.51.0 H L6R30.00.51.0 0.00.51.0 H L7P30.00.51.0 0.00.51.0 H L7R30.00.51.0 0.00.51.0 H L8P30.00.51.0 0.00.51.0 H L8R30.00.51.0 0.00.51.0 H L9P30.00.51.0 0.00.51.0 H L9R30.00.51.0 0.00.51.0 H L10P30.00.51.0 0.00.51.0 H L10R30.00.51.0 0.00.51.0 H W0R30.00.51.0 0.00.51.0 H W0Q30.00.51.0 −200 −150 −100 −50 0 50 100Relative velocity (km s −1 )0.00.51.0 H W1R30.00.51.0 −200 −150 −100 −50 0 50 100Relative velocity (km s −1 )0.00.51.0 H W1Q30.00.51.0
Fig. 19.
Fit to H (J =
3) lines.
16. Noterdaeme et al.: A translucent interstellar cloud at z = . L0P40.500.751.00 0.00.51.0 H L0R40.00.51.0 0.00.51.0 H L1P40.00.51.0 0.00.51.0 H L1R40.00.51.0 0.00.51.0 H L2P40.00.51.0 0.00.51.0 H L2R40.00.51.0 0.00.51.0 H L3P40.00.51.0 0.00.51.0 H L3R40.00.51.0 0.00.51.0 H L4P40.00.51.0 0.00.51.0 H L4R40.00.51.0 0.00.51.0 H L5P40.00.51.0 0.00.51.0 H L5R40.00.51.0 0.00.51.0 H L6P40.00.51.0 0.00.51.0 H L6R40.00.51.0 0.00.51.0 H L7P40.00.51.0 0.00.51.0 H L7R40.00.51.0 0.00.51.0 H L8P40.00.51.0 0.00.51.0 H L8R40.00.51.0 0.00.51.0 H L9P40.00.51.0 0.00.51.0 H L9R40.00.51.0 0.00.51.0 H L10P40.00.51.0 0.00.51.0 H L10R40.00.51.0 0.00.51.0 H W0R40.00.51.0 0.00.51.0 H W0Q40.00.51.0 −200 −150 −100 −50 0 50 100Relative velocity (km s −1 )0.00.51.0 H W1R40.00.51.0 −200 −150 −100 −50 0 50 100Relative velocity (km s −1 )0.00.51.0 H W1Q40.00.51.0
Fig. 20.
Fit to H (J =
4) lines. L0P50.80.91.0 0.500.751.00 H L0R50.500.751.00 0.00.51.0 H L1P50.00.51.0 0.00.51.0 H L1R50.00.51.0 0.00.51.0 H L2P50.00.51.0 0.00.51.0 H L2R50.00.51.0 0.00.51.0 H L3P50.00.51.0 0.00.51.0 H L3R50.00.51.0 0.00.51.0 H L4P50.00.51.0 0.00.51.0 H L4R50.00.51.0 0.00.51.0 H L5P50.00.51.0 0.00.51.0 H L5R50.00.51.0 0.00.51.0 H L6P50.00.51.0 0.00.51.0 H L6R50.00.51.0 0.00.51.0 H L7P50.00.51.0 0.00.51.0 H L7R50.00.51.0 0.00.51.0 H L8P50.00.51.0 0.00.51.0 H L8R50.00.51.0 0.00.51.0 H L9P50.00.51.0 0.00.51.0 H L9R50.00.51.0 0.00.51.0 H L10P50.00.51.0 0.00.51.0 H L10R50.00.51.0 0.00.51.0 H W0R50.00.51.0 0.00.51.0 H W0Q50.00.51.0 −200 −150 −100 −50 0 50 100Relative velocity (km s −1 )0.00.51.0 H W1R50.00.51.0 −200 −150 −100 −50 0 50 100Relative velocity (km s −1 )0.00.51.0 H W1Q50.00.51.0
Fig. 21.
Fit to H (J =
5) lines.5) lines.