Statistics and characteristics of MgII absorbers along GRB lines of sight observed with VLT-UVES
Susanna D. Vergani, Patrick Petitjean, Cedric Ledoux, Paul Vreeswijk, Alain Smette, Evert J.A. Meurs
AAstronomy & Astrophysics manuscript no. MgIIGRB c (cid:13)
ESO 2018November 2, 2018
Statistics and characteristics of MgII absorbers along GRB lines ofsight observed with VLT-UVES
Susanna D. Vergani , , , Patrick Petitjean , C´edric Ledoux , Paul Vreeswijk , Alain Smette , and Evert J.A. Meurs , University Paris 7, APC, Lab. Astroparticule et Cosmologie, UMR7164 CNRS, 10 rue Alice Domon et Lonie Duquet, F-75205,Paris Cedex 13, France University Paris 6, Institut d’Astrophysique de Paris, UMR7095 CNRS, 98bis Boulevard Arago, F-75014, Paris, France School of Physical Sciences and NCPST, Dublin City University, Dublin 9, Ireland European Southern Observatory, Alonso de C´ordova 3107, Casilla 19001, Vitacura, Santiago 19, Chile Dark Cosmology Centre, Niels Bohr Institute, University of Copenhagen, DK-2100 Copenhagen, Denmark School of Cosmic Physiscs, DIAS, 31 Fitzwilliam Street, Dublin 2, IrelandReceived date / Accepted date
ABSTRACT
We analyse the properties of Mg ii absorption systems detected along the sightlines toward GRBs using a sample of 10 GRB afterglowspectra obtained with VLT-UVES over the past six years. The signal-to-noise ratio is su ffi ciently high that we can extend previousstudies to smaller equivalent widths (typically W r > ∆ z ∼
14, we detect 9 intervening Mg ii systemswith W r > < W r < σ significance. Using intermediate and low resolution observations reported in the literature,we increase the absorption length for strong systems to ∆ z = ii systems is a factor of 2.1 ± σ significance) toward GRBs as compared to QSOs,about twice smaller however than previously reported. We divide the sample in three redshift bins and we find that the number densityof strong Mg ii is larger in the low redshift bins. We investigate in detail the properties of strong Mg ii systems observed with UVES,deriving an estimate of both the H i column density and the associated extinction. Both the estimated dust extinction in strong GRBMg ii systems and the equivalent width distribution are consistent with what is observed for standard QSO systems. We find also thatthe number density of (sub)-DLAs per unit redshift in the UVES sample is probably twice larger than what is expected from QSOsightlines which confirms the peculiarity of GRB lines of sight. These results indicate that neither a dust extinction bias nor di ff erentbeam sizes of the sources are viable explanations for the excess. It is still possible that the current sample of GRB lines of sight isbiased by a subtle gravitational lensing e ff ect. More data and larger samples are needed to test this hypothesis. Key words. quasars: absorption lines – gamma rays: bursts
1. Introduction
Thanks to their exceptional brightness, and although fading veryrapidly, Gamma-Ray Burst (GRB) afterglows can be used aspowerful extragalactic background sources. Since GRBs can bedetected up to very high redshifts (Greiner et al. 2008; Kawaiet al. 2006; Haislip et al. 2006) their afterglow spectra can beused to study the properties and evolution of galaxies and theIGM, similarly to what is traditionally done using QSO spectra.Even if the number of available GRB lines of sight (los) ismuch smaller than those of QSOs, it is interesting to compare thetwo types of lines of sight. In particular, Prochter et al. (2006b)found that the number density of strong (rest equivalent width W r > ii absorbers is more than 4 timeslarger along GRB los than what is expected for QSOs over thesame path length. This result has been derived from a sample of14 GRB los and a redshift path of ∆ z = .
5, and has been con-firmed by Sudilovsky et al. (2007). Dust extinction bias for QSOlos, gravitational lensing, contamination from high-velocity sys-tems local to the GRB and di ff erence of beam sizes are amongthe possible causes of this discrepancy. All these e ff ects can con-tribute to the observed excess, but no convincing explanation has Send o ff print requests to : S.D. Vergani, [email protected] been found to date for the amplitude of the excess (Prochter et al.2006b; Frank et al. 2007; Porciani et al. 2007). Similar studies(Sudilovsky et al. 2007; Tejos et al. 2007) have been performedon the number of C iv intervening systems. Their results are inagreement with QSO statistics.Clearly, further investigation of this excess is required. Sincethe reports by Prochter et al. (2006b) (based on a inhomoge-neous mix of spectra from the literature) and the confirmationby Sudilovsky et al. (2007) (based on a homogeneous but lim-ited sample of UVES los), several new los have been observed.As of June 2008, the number of los with good signal-to-noiseratio observed by UVES has increased to 10. We use this sampleto investigate the excess of strong Mg ii absorbers and, thanks tothe high quality of the data, we also extend the search of systemsto lower equivalent widths and derive physical properties of theabsorbing systems. In addition, to increase the redshift path overwhich strong Mg ii systems are observed, we consider also a sec-ond sample that includes in addition other observations gatheredfrom the literature.We describe the data in Section 2, identify Mg ii systems anddetermine their number density in Section 3. We derive somecharacteristics of strong Mg ii systems in Section 4, we estimatetheir HI content in Section 5 and we study peculiar (sub-)DLAs a r X i v : . [ a s t r o - ph . H E ] J un Susanna D. Vergani et al.: MgII absorbers along GRB lines of sight observed with UVES systems detected along the lines of sight in Section 6. We sum-marize and conclude in Section 7.
2. Data
Our first sample (herafter the
UVES sample ) includes tenGRB afterglows with available follow-up VLT / UVES high-resolution optical spectroscopy as of June 2008: GRB 021004,GRB 050730, GRB 050820A, GRB 050922C, GRB 060418,GRB 060607A, GRB 071031, GRB 080310, GRB 080319B andGRB 080413A. All GRBs were detected by the Swift satellite(Gehrels et al. 2004), with the exception of GRB 021004, whichwas detected by the
High-Energy Transient Explorer (HETE-2) satellite (Ricker et al. 2003).UVES observations began on each GRB afterglow with theminimum possible time delay. Depending on whether the GRBlocation was immediately observable from Paranal, and whetherUVES was observing at the time of the GRB explosion, the af-terglows were observed in either Rapid-Response Mode (RRM)or as Target-of-Opportunity (ToO). A log of the observations isgiven in Table 1.
Table 1.
GRB sample and log of UVES observations
GRB UT a δ t b t total c ESO Program PI(yymmdd)
Swift (hh:mm) (h) ID021004 12:06:13 13:31 2.0 070.A-0599 d Fiore050730 19:58:23 04:09 1.7 075.A-0603 Fiore050820A 06:34:53 00:33 1.7 075.A-0385 Vreeswijk050922C 19:55:50 03:33 1.7 075.A-0603 Fiore060418 03:06:08 00:10 2.6 077.D-0661 Vreeswijk060607A 05:12:13 00:08 3.3 077.D-0661 Vreeswijk071031 01:06:36 00:09 2.6 080.D-0526 Vreeswijk080310 08:37:58 00:13 1.3 080.D-0526 Vreeswijk080319B 06:12:49 00:09 2.1 080.D-0526 e Vreeswijk080413A 02:54:19 03:42 2.3 081.A-0856 Vreeswijk a UT of trigger by the BAT instrument on-board
Swift . Exception:GRB 021004, detected by
WXM on-board
HETE-2 . b Time delay between the satellite trigger and the start of the firstUVES exposure: normally a series of spectra is taken. c Total UVES exposure time including all instrument setups. d Also 070.D-0523 (PI: van den Heuvel). e Also 080.A-0398 (PI: Fiore).
The observations were performed with a 1.0 (cid:48)(cid:48) wide slitand 2x2 binning, providing a resolving power of R ≈
48 000(FWHM ∼ − ) for a ≈ − pixel size . The UVESdata were reduced with a customized version of the MIDAS re-duction pipeline (Ballester et al. 2006). The individual scientificexposures were co-added by weighting them according to the in-verse of the variance as a function of wavelength and rebinnedin the heliocentric rest frame.Although the UVES sample has a smaller number of loscompared to the sample used by Prochter et al. (2006b) (10 in-stead of 14 los), the redshift path of the two samples is simi-lar ( ∆ z = UVES is described in Dekker et al. (2000). Though the minimum guaranteed resolving power of UVES in thismode is 43 000, we find that in some cases a higher resolution, up to ≈
50 000, is achieved in practice, due to variations in seeing conditions. ∆ z = ii statistics.The second sample we consider (the overall sample ) isformed by adding observations from the literature (see Table 4)to the UVES sample (see Section 3.2). The sample gathers ob-servations of 26 GRBs for ∆ z =
3. Number density of Mg ii absorbers For each line of sight we searched by eye the spectrum for Mg ii absorbers outside the Lyman- α forest considering all Mg ii com-ponents within 500 km s − as a single system. Table 2 summa-rizes the results. Columns 1 to 8 give, respectively, the name ofthe GRB, its redshift, the redshift paths along the line of sightfor W r , lim > ii ab-sorber, the rest equivalent width of the Mg ii λ ii λ ii doublet and the contaminating absorption and derivecharacteristics of the Mg ii λ W r , lim (cid:39) U M . S NR ∆ λ (Å) , (1)where M L is the number of pixels the line is expected to cover, U is the number of rms-intervals (or σ ) defining the statisticalsignificance of the detection limit, S NR is the signal-to-noiseratio at the corresponding wavelength and ∆ λ is the FWHM ofthe resolution element. We will apply this detection limit to theMg ii λ M L = U = ii doublet is detected at more than 3 σ ), it canbe seen that for a typical signal-to-noise ratio of SNR ∼
10 anda typical FWHM of ∆ λ = R = λ = ∼ . − z max ∼ z = z start = . z max = usanna D. Vergani et al.: MgII absorbers along GRB lines of sight observed with UVES 3 Table 2.
Characteristics of the Mg ii absorbers in the UVES sample. GRB z GRB ∆ z ∆ z z abs W r ( λ ej RemarksW r > r >
1Å (Å) (km / s)021004 2.3295 1.754 1.756 0.5550 0 . ± . ∼ z = . . ± . ∼ . ± . ∼ . ± . ∼ < .
783 (0 . ∼ . ± . ∼ . ± . ∼ . ± . ∼ . ± . ∼ < .
424 (0 . ∼ z = . . ± . ∼ . ± . ∼ < .
102 (0 . ∼ . ± . ∼ . ± . ∼ . ± . ∼ . ± . ∼ . ± . ∼ . ± . ∼ . ± . ∼ . ± . ∼ . ± . ∼ . ± . ∼ . ± . ∼ . ± . ∼ . ± . ∼ . ± . ∼ . ± . ∼ . ± . ∼ km s − from the GRB redshift. Table 3 lists the mean redshift, (cid:104) z abs (cid:105) , the total redshift paths obtained considering these red-shift limits, and the number of systems detected in our sam-ple over these redshift paths for di ff erent W r limits or ranges: W r > . ’all systems’ ), W r > . ’strong systems’ ) and0 . ≤ W r ≤ . ’weak systems’ ). The total redshift pathsfor the all and strong samples are ∆ z = .
79 and 13 .
94, re-spectively, for an observed number, N MgIIobs , of 18 and 9 sys-tems detected, corresponding to redshift number densities of ∂ n /∂ z = . ± .
31, 0.65 ± all and strong Mg ii sys-tems, respectively. Fig. 1.
Redshift path density g ( z ) of the UVES sample for W r > . W r > . Table 3.
Number of Mg ii systems and redshift paths W r ( λ > > > . < (cid:104) z abs (cid:105) N MgIIobs (UVES sample) 18 9 9 N MgIIexp .
98 ( ± .
46) 4 .
83 ( ± .
20) 7 .
21 ( ± . N MgIIexp .
00 ( ± . In Table 3 we also report the number of Mg ii absorbers thatwould be expected along lines of sight toward QSOs over thesame redshift path, N MgIIexp . To calculate these numbers, we deter-mine the redshift path density, g ( z ), of the UVES GRB sample(see Fig. 1) and combine it with the density of QSO absorbers Susanna D. Vergani et al.: MgII absorbers along GRB lines of sight observed with UVES
Fig. 2.
Comparison between the cumulative distribution of Mg ii systems detected along the UVES GRB los (red) and the oneexpected along QSO los following Eq. 2 (black solid) or Eq. 3(black dashed), for: all (top panel), strong (middle panel) and weak (bottom panel) systems.per unit redshift as observed by Nestor et al. (2005) and Prochteret al. (2006a) (for the strong systems only) in the SDSS survey: N MgIIexp = (cid:90) z end z start g ( z ) ∂ n ∂ z dz . (2)Both Nestor et al. (2005) and Prochter et al. (2006a) showedthat the traditional parametrization of ∂ n /∂ z as a simple pow-erlaw n (1 + z ) γ does not provide a good fit to the SDSS data.Therefore we will use their empirical fits to the redshift numberdensity. Nestor et al. (2005) give ∂ n ∂ z = (cid:90) dndW r ( z ) dW = (cid:90) n ∗ ( z ) W ∗ ( z ) e − W r / W ∗ dW , (3)where both n ∗ and W ∗ vary with redshift as power laws: n ∗ = . ± . + z ) . ± . , W ∗ = . ± . + z ) . ± . .Prochter et al. (2006b) derive for the strong MgII systems: ∂ n /∂ z = − . + . z − . z + . z (4)The results of the calculations for the di ff erent W r ranges( W r > . W r > . . ≤ W r ≤ . (cid:113) N MgIIexp .As a verification, we note from Fig. 13 of Nestor et al. (2005)that the number density, ∂ n /∂ z , of W r > . ∂ n /∂ z = .
783 valuereported there by the UVES ’all systems’ ∆ z = .
79 we obtainan expected total number of 9.78, in agreement with the valuereported in Tab. 3 for systems along QSO los (i.e. 10.54). Thesame test performed using the number density values found bySteidel & Sargent (1992) give consistent results.Fig. 2 and Table 3 show that the excess of strong Mg ii sys-tems along GRB los compared to QSO los is significant at morethan 2 σ (slightly less than 2 σ for the strong Mg ii systems if theNestor et al. 2005 function is used), but it is more than a fac-tor of ∼ ∼ σ to thatexpected for QSO los. Fig. 3.
Comparison between the cumulative distribution ofstrong Mg ii systems in the overall sample of GRB los (red) andthe one expected along QSO los following Eq. 2 (black solid) orEq. 3 (black dashed).In order to increase the statistics of strong Mg ii systems, weadded to the UVES sample both high and low resolution GRB af-terglow spectra published in the literature. The resulting sampleis composed of 26 los (see Table 4 for details on los that are notpart of the UVES sample). Since this sample includes many lowresolution spectra, we study the statistics of strong ( W r > ii absorbers only, using the same redshift limits as in the pre-vious Section. Three lines of sight of this sample (GRB 991216,GRB 000926 and GRB 030429) were not used by Prochter et al.(2006b) because the low spectral resolution of these spectra doesnot allow to resolve the Mg ii doublet. However, such a low reso-lution does not prevent the detection of strong systems, althoughthe doublet is blended. In addition, the total equivalent widthof the doublets detected along these los is larger than 3 Å sothat we are confident that W r , is larger than 1 Å. In any case,as detection of Mg ii systems is more di ffi cult at lower resolu-tion, including these lines of sight could only underestimate theirnumber density.The total number of strong Mg ii systems is N =
22 andthe redshift path is ∆ z = .
55. This leads to a number density ∂ n /∂ z = . ± .
15. We use the g ( z ) function of this enlarged usanna D. Vergani et al.: MgII absorbers along GRB lines of sight observed with UVES 5 Table 4.
GRB los available from the literature.
GRB z GRB z start z end z abs W r (2796) ∆ v ej Reference † (Å) (km / s)991216 1.022 0.366 1.002 0.770 a . ± . b ∼ . ± . b ∼ . ± . ∼ . ± . ∼ . ± . ∼ . ± . ∼ . ± . b ∼ . ∼ . ∼ . ± . ∼ . ± . ∼ . ∼ . ± . ∼ †
1: Castro et al. (2003); 2: Mirabal et al. (2002); 3: Vreeswijk et al. (2006); 5: Barth et al. (2003); 4: Masetti et al. (2003);6: Klose et al. (2004); 7: Vreeswijk et al. (2004); 8: Berger et al. (2006); 9: Prochter et al. (2006b); 10: Chen et al. (2008);11: Th¨one et al. (2008); 12: Perley et al. (2008); 13: Hao et al. (2007); 14: Maiorano et al. (2006); 15: Jakobsson et al. (2004). a The very low resolution of the GRB 991216 spectrum makes the z = b The W r values refer to the totalequivalent width of the MgII doublet. sample to compute the total number of strong systems expectedfor a similar QSO sample following the same method as used inSect. 3.1. We find N MgIIexp = . ± .
23 and N MgIIexp = . ± . . ± . . ± . ii systems along GRB lines of sightfor this enlarged sample is confirmed at a ∼ σ statistical sig-nificance. The excess found is higher than for the UVES samplebut still a factor of ∼ ff erent by no more than 2 σ . For consistency we also per-formed our analysis considering only the smaller sample usedby Prochter et al. (2006b). In this case, the results obtained aresimilar to those found by these authors. The redshift path of ouroverall sample is twice as large as that used by Prochter et al.(2006b), therefore the factor of 2 excess we find in this studyhas a higher statistical significance. We divided both the UVES and the overall sample in three red-shift bins and calculated ∂ n /∂ z for each bin. Fig. 4 shows thenumber of systems per redshift bin. While the total number ofsystems and the number of weak systems have a comparableredshift evolution in GRB and QSO lines of sight, the strongsystems happen to have a di ff erent evolution in GRB los. Theexcess of strong systems in GRBs is particularly pronounced atlow redshift, up to z ∼ .
6. We performed a KS test for each ofthe three cases to assess the similarity of the redshift distribu-tion of Mg ii systems along GRB and QSO los. There is a 90 . .
5% chance that the weak and all Mg ii absorber sam-ples in QSOs and GRBs are drawn from the same population.The probability for the strong systems is 20 .
1% for the UVES sample, but it decreases to ∼
2% when considering the overallsample.This apparent excess of strong systems in the low-redshiftbin could indicate that some amplification bias due to lensing isat work. Indeed the e ff ect of lensing should be larger in casethe deflecting mass is at smaller redshift. The lensing opticaldepth for GRBs at redshift z GRB >> z GRB ≤ z l ∼ . z GRB /
2. However, wefind that only 47% of GRBs with strong foreground absorbersin our sample have at least one strong Mg ii system locatedin the range within which the optical depth decreases to abouthalf its maximum value (Sudilovsky et al. 2007; Turner 1980;Smette et al. 1997). This indicates that if amplification by lens-ing is the correct explanation, the e ff ect must be weak and subtle.Porciani et al. (2007) used the optical afterglow luminosities re-ported by Nardini et al. (2006) to show that GRB afterglows withmore than one absorbers are brigther than others by a factor 1.7.However this correlation has not been detected by Sudilovskyet al. (2007) when using the afterglow B-band absolute magni-tudes obtained by Kann et al. (2006, 2007).More data and larger samples are obviously needed to con-clude on this issue.
4. The population of strong Mg ii systems Both the results on the UVES homogeneous sample and thoseobtained using the overall sample confirm, although to a smallerextent, the excess of strong Mg ii absorbers along GRB los firstreported by Prochter et al. (2006b). To understand the reason ofthe discrepancy in the number density of “strong” systems it istherefore important to study in more details these systems andto derive their physical characteristics. This is possible using theUVES data. The main question we would like to address here iswhether there is any reason to believe that GRB and QSO strongabsorbers are not drawn from the same population. Susanna D. Vergani et al.: MgII absorbers along GRB lines of sight observed with UVES
Fig. 4.
Number density evolution of Mg ii systems detectedalong GRB los for: first panel: all , second panel: strong , thirdpanel: weak systems of the UVES sample ; bottom panel: strong systems of the overall sample . The solid line represents the evo-lution of the MgII number density derived from the Mg ii surveyin the SDSS QSO by Nestor et al. (2005) (see Eq. 3). We show in Fig. 5 the comparison between the normalized W r distribution of all Mg ii systems with W r > UVES sample and the one reported by Nestor et al. (2005) forthe Mg ii systems along the QSO los in the SDSS survey.The KS tests give a 27 .
1% chance that the GRB and QSOdistributions are drawn from the same population for the
UVESsample . This result extends to lower W r values the conclusionby Porciani et al. (2007) that the two distributions are similar,arguing against the idea that the excess of Mg ii systems couldbe related to the internal structure of the intervening clouds. Fig. 5. W r distribution of the Mg ii systems with W r > Fig. 6.
The W r > . ii absorption systems detected in theUVES spectra. the Mg ii λ λ − centered at the redshift reported in Table1. ii systems The profiles of the two transitions of the W r > ii systemsfound in the UVES spectra are plotted on a velocity scale inFig. 6. The profiles are complex, spread over at least 200 km s − and up to ∼
600 km s − and show a highly clumpy structure.We measure the velocity spread of the Mg ii systems with W r > UVES sample following Ledoux et al. (2006).We therefore use a moderately saturated low ionization absorp- usanna D. Vergani et al.: MgII absorbers along GRB lines of sight observed with UVES 7
Fig. 7.
The solid line histogram represents the distribution of thevelocities of the strong Mg ii systems in the UVES sample cal-culated following Ledoux et al. (2006). The thick dashed his-togram refers to the 27 SDSS Mg ii systems observed at highspectral resolution (Prochter et al. 2006a).tion line (e.g. FeII λ λ ff erence between the points of the absorption profiles at which5% and 95% of the absorption occurs. This method is definedin order to measure the velocity width of the bulk of gas, avoid-ing contamination by satellite components which have negligi-ble contribution to the total metal column density. Note that us-ing this definition implies that the measured velocity spread isusually smaller than the spread of the Mg ii profile which isstrongly saturated. In case no moderately saturated absorptionline is available, we use the mean value of the velocity widthscalculated both from a saturated line and an optically thin line.Results are given in column 9 of Table 6.The velocity-spread distribution is shown for UVES systemswith W r > W r > It has been proposed that a dust bias could possibly a ff ect thestatistics of strong Mg ii systems. Indeed, if part of the popu-lation of strong Mg ii systems contains a substantial amount ofdust, then the corresponding lines of sight could be missed inQSO surveys because of the attenuation of the quasar whereasGRBs being intrinsically brighter, the same lines of sight arenot missed when observing GRBs. The existence of a dust ex-tinction bias in DLA surveys is still a debated topic among theQSO community although observations of radio selected QSOlos (Ellison et al. 2004) seem to show that, if any, this e ff ect isprobably small (see also Pontzen & Pettini 2008).We can use our UVES lines of sight to estimate the dust con-tent of strong Mg ii systems from the depletion of iron com-pared to other non-depleted species as is usually done in DLAs.We have therefore searched for both Fe ii and Zn ii absorptionlines (or Si ii , in case the Zn ii lines are not available) associ-ated to the strongest Mg ii systems present in the UVES spec-tra. Because the spectra do not always cover the relevant wave- length range, the associated column densities could be measuredonly for 4 out of the 10 systems (see Table 5). We estimate thedepletion factor, and therefore the presence of dust, from themetallicity ratio of iron (a species heavily depleted into dustgrains in the ISM of our Galaxy) to zinc (that is little depleted),[Fe / Zn] = (Fe / Zn) − (Fe / Zn) (cid:12) (or [Fe / Si]). We also determine theiron dust phase column density ( N dustFe ) using the formula givenby Vladilo et al. (2006) and from this we infer an upper limiton the corresponding flux attenuation A V from their Fig. 4. Weused also the estimate given by (Bohlin et al. 1978; Prochaska &Wolfe 2002): A V = . N (HI)10 κ = . X × [X / H] (cid:12) (1 − [Fe / X] ) (5)with κ = [X / H] (1 − [Fe / X] ) representing the dust-to-gasratio and X corresponding to Zn or Si if Zn is not available.Results are reported in Table 5. It can be seen that al-though depletion of iron can be significant, the correspondingattenuation is modest because the column densities of metal-lic species are relatively small owing to low metallicities (seealso Table 5). Indeed we find high dust depletion in two sys-tems at z abs = z abs = A V doesnot exceed values typically found for QSO los (see Vladiloet al. 2006 and Prantzos & Boissier 2000). Cucchiara et al.(2008) do not detect the Zn II absorption lines for the systemsat z abs = W r < .
016 Å whereas we find W r = .
07 Å. The detectionof Zn II( λλ λλ ii systems identifiedalong the los. Therefore our results do not show evidences tosupport the idea that a bias due to the presence of dust in strongMg ii systems could be an explanation for the overabundance ofthe strong Mg ii absorbers along GRB los. On the other hand, thefact that 2 out of 4 of the selected systems have a dust depletionhigher than the values usually found for QSO DLAs (Meiringet al. 2006) supports the fact that these systems probably arisein the central part of massive halos where the probability to findcold gas is expected to be higher.
5. Estimating the H i column density A key parameter to characterize an absorber is the correspondingH i column density. Unfortunately, the H i Lyman- α absorptionline of most of the systems is located below the atmosphericcut-o ff and is unobservable from the ground. If we want to char-acterize the systems with their H i column density or at least anestimate of it, we have to infer it indirectly. For this we will as-sume that the strong systems seen in front of GRBs and QSOsare cosmological and drawn from the same population. We be-lieve that the results discussed in the previous section justify thisassumption.We use the velocity-metallicity correlation found by Ledouxet al. (2006) to estimate the expected metallicity of the systemsin the UVES sample with W r > W r > (cid:104) z (cid:105) = .
11, giving
Susanna D. Vergani et al.: MgII absorbers along GRB lines of sight observed with UVES
Table 5.
Iron to Zinc or Silicon ratio and extinction estimate for 4 strong Mg ii systems. z N (FeII) N (ZnII) N (SiII) [Fe / Zn] [Fe / Si] N dustFe A V (cm − ) (cm − ) (cm − ) (cm − )GRB021004 1.3800 15 . ± .
05 15 . ± . − . ± .
10 14 . ± . ∼ . ± .
03 12 . ± . − . ± .
04 15 . ± . < . . ± .
01 12 . ± . − . ± .
03 15 . ± . < . . ± .
03 14 . ± . − . ± .
10 13 . ± . ∼ Table 6.
Characteristics of Mg ii systems with W r > . GRB z abs W MgII λ W MgI λ W FeII λ N ZnII N SiII N FeII a ∆ v b ∆ v MgII λ (Å) (Å) (Å) cm − cm − cm − km s − km s − . ± .
09 15 . ± .
05 170 89021004 1.603 1.407 0.366 0.737 12 . ± .
02 14 . ± .
03 164 51050820A 0.692 2.874 N / A c N / A c N / A c . ± .
03 271 98060418 0.603 1.293 0.361 0.989 16 . ± . d
76 130060418 0.656 1.033 0.078 0.486 13 . ± .
02 136 123060418 1.107 1.844 0.483 1.080 12 . ± .
03 14 . ± .
01 119 221060607A 1.803 1.854 > .
226 0.825 14 . ± . . ± .
03 333 90080319B 0.715 1.482 0.303 0.697 14 . ± .
01 354 142 a Velocities are measured following Ledoux et al. (2006); b Velocities are measured using the MgII λ c Lines redshifted on top of the Ly − α absorption associated to theGRB or in the Ly − α forest; d Saturated line. [X / H] = . ∆ v − .
78. We infer the hydrogen column densi-ties dividing the zinc, silicon or iron column densities measuredin the UVES spectrum by the metallicity. Note that in most ofthe cases only Fe ii available (see Table 5). The N (H i ) columndensity derived in these cases should be considered a lower limitbecause iron can be depleted onto dust-grains. The results areshown in Table 7 (columns 3 and 4). The error on the proce-dure should be of the order of 0.5 dex (see Ledoux et al. 2006).We insist on the fact that our aim is not to derive an exact H i column density for each system but rather to estimate the over-all nature of the systems. We find that at least 3 of the 9 sys-tems with W r > N (H i ) > ii absorbers is to use the ‘ D -index’ (Ellison et al. 2009). We cal-culate the D -index for the systems in our sample following therecommended D -index definition by Ellison et al. (2009) and us-ing the formula: D = W r (MgII λ ∆ v (MgII λ / s) × log N (FeII)15 × . (6)The width of the central part of the Mg ii λ ∆ v (MgII λ / s, is reported in column 10 ofTable 6 while the resulting D -index values are reported in Table 7 (column 5). Note that if we do not include the iron column den-sity term in the D -index calculation, we obtain similar results.We find that all the 9 systems have D >
7. Ellison et al.(2009) find that 57 + . − . % of the systems having D > n DLA ) found for the QSO los by Rao et al. (2006) to that ofour GRB sample. n DLA is defined as the product of the Mg ii sys-tems number density and the fraction of DLAs in a Mg ii sample.Rao et al. (2006) found n DLA , QSO ∼ . n DLA , GRB ∼ . − .
36 . This means that thenumber of DLAs is at least two times larger along GRB lines ofsight as compared to QSO lines of sight.The above estimate of log N (H i ) can be considered as highlyuncertain and the identification of a few of the systems as DLAscan be questioned.All this seems to indicate that GRBs favor lines of sight withan excess of DLAs. Since these systems are more likely to belocated in the central parts of massive halos, this may again favorthe idea that there exists a bias in GRB observations towardsGRBs with afterglows brighter because they are subject to somelensing amplification. usanna D. Vergani et al.: MgII absorbers along GRB lines of sight observed with UVES 9 Fig. 8.
Fe II( λλ λλ λλ z abs = z abs = Table 7.
Inferred metallicity, N HI and D -index of Mg ii systemswith W r > . GRB z abs [X / H] a N HI (cm − ) b D -index c − .
63 20.3 8 . ± . − .
65 20.8 9 . ± . . ± . d − . > . . ± . − . > . . ± . e − . > . . ± . − .
85 21.1 8 . ± . − .
21 19.0 9 . ± . − . > . . ± . a Metallicities inferred using the velocity-metallicity relation found byLedoux et al. (2006); b N HI values inferred using the velocity-metallicityrelation found by Ledoux et al. (2006). When only the iron column den-sity is available, the N HI value has to be considered has a lower limit dueto possible dust extinction; c D -index calculated using Eq. 6, followingthe method recommended by Ellison et al. (2009); d D -index calculatedwithout including the iron column density term (see Ellison et al. 2009); e iron lines are saturated. The D -index calculated without including theiron column density term would be D = . ± .
6. Observed sub-DLA absorbers
The UVES spectra often cover a substantial part of the Lyman- α forest in front of the GRB. It is therefore possible to searchdirectly for strong H i Lyman- α absorption lines correspond-ing to DLAs or, more generally, sub-DLAs. This can onlybe performed for 8 of the 10 GRBs in the sample: the red-shifts of GRB 060418 and GRB 080319B are unfortunately toolow to allow the detection of the Lyman- α absorptions in the UVES spectral range. We exclude from our search the DLA atthe GRB redshift, which is believed to be associated with theclose surrounding of the GRB. We find additional (sub-)DLAs(Fig. 9) along the los of GRB 050730 (see also Chen et al. 2005),GRB 050820, GRB 050922C (see also Piranomonte et al. 2008)and GRB 060607A (Fig. 10). The former three systems are sim-ple, with a single Lyman- α component. The system detected at2.9374 toward GRB 060607A is more complex and cannot be fitwith a single component (sub-) DLA. The profile is made of twomain clumps at z = . i components (see top panel in Fig. 9). The constraintson the H i column densities come mostly from the red wing ofthe Lyman- α line and the structures seen in the Lyman- γ line.We find that the two main H i components at z = . N (H i ) = . . ± .
1, respectively. It isworth noting that the limit on the O i component at z = . N (O i ) < . / H] < − . z start as the redshift of an absorber, for which the correspondingLyman- α line would be redshifted to the same wavelength as theLyman- β line of the GRB afterglow. We therefore avoid confu-sion with the Lyman- β forest. z end is fixed at 3000 km / s from theGRB redshift. The only exception to this rule is GRB 050730 forwhich there is a gap in the spectrum located at about 3000 km / sfrom the GRB redshift and starting at λ = z max . The corresponding total redshift path is ∆ z = .
17. The sub-DLAs systems at z = . ∼
400 km s − , thereforefor the statistical study we consider them as a single system. Theresulting number density for systems with log N (H i ) > . ± .
48 for an average redshift of (cid:104) z (cid:105) = .
08. At this redshifta value of about 0 . ± . α forest closest to the GRB, with ejec-tion velocities of about 25000, 22000, 12000 and 11000 km s − for GRB 050730, 050820A, 050922C and 060607A respec-tively. The corresponding ejection velocities found for the strongMg ii systems (see column 7 of Table 2) are larger than35000 km / s, making an origin local to the GRB unlikely for thestrong Mg ii systems. This fact therefore does not favor the pos-sibility that the excess of strong Mg ii absorbers along GRB losis due to some ejected material present in the GRB environment.The much lower ejection velocities found for the sub-DLA ab-sorbers may indicate that for these systems the excess is not ofthe same origin as the excess of Mg ii systems and that part ofthe gas in these systems could have been ejected by the GRB.
7. Conclusion
We have taken advantage of UVES observations of GRB after-glows obtained over the past six years to build an homogeneous
Fig. 9.
Foreground (sub)-DLAs.
Top panel : Sub-damped Lyman- α absorption at z = . Middlepanel : Damped Lyman- α absorption at z = . Bottom panel : Sub-damped Lyman- α absorptionat z = . z = . > ii systems alongGRB lines of sight extending the study to smaller equivalentwidths. We also used these data to derive intrinsic physical prop-erties of these systems.Considering the redshift ranges 0 . < z < .
27 of the SDSSsurvey used for the QSO statistics, we find an excess of strongintervening MgII systems along the 10 GRB lines of sight ob-served by UVES of a factor of ∼ ∼ σ . Thanks to the quality ofthe UVES data it has been possible also to consider the statisticsof the weak absorbers (0.3 < W r < ∆ z = ∼ σ significance. Fig. 10.
The sub-DLA system at z = . i λ γ , Lyman- β and Lyman- α absorption lines(from top to bottom, respectively). The red lines show the main z = . ff erent lines include errors) mostlyconstrained by the red wing of the Lyman- α absorption. Inblue is shown the total fit including the sub-DLA componentat z = . ff erent at less than 2 σ . Thepresent result is statistically more significant.Possible explanations of this excess include: dust obscura-tion that could yield such lines of sight to be missed in quasarstudies; di ff erence of the beam sizes of the two types of back-ground sources; gravitational lensing. In order to retrieve moreinformation to test these hypotheses we investigate in detail theproperties of the strong Mg ii systems observed with UVES. Wefind that the equivalent width distribution of Mg ii systems issimilar in GRBs and QSOs. This suggests that the absorbers are usanna D. Vergani et al.: MgII absorbers along GRB lines of sight observed with UVES 11 Table 8.
Properties of the foreground damped Ly- α systems detected along the UVES GRB spectra. ∆ z a z log N HI log N OI log N SiII log N FeII [Fe / Si] [O / H] [Si / H] [Fe / H](Sub-)DLA ( N in cm − ) ( N in cm − ) ( N in cm − ) ( N in cm − )GRB021004 0.487 There are no intervening (Sub-)DLA along this losGRB050730 0.542 3.5655 20 . ± . N / A b < . < . < − . < − . . ± . N / A b . ± .
02 14 . ± . − . ± . − . ± . − . ± . . ± . > .
30 14 . ± .
10 14 . ± .
03 0 . ± . > − . − . ± . − . ± . . c . ± . > . > . d . ± . < − . > − . > − . − . ± . . c . ± . < . < . < . < − . < − . < − . a Redshift path calculated from the Ly- β GRB absorption line to 3000 km / s from the GRB Ly- α , except for GRB 050730 where the z range ends at the beginning of the spectral gap at λ = b The line is blended with other lines in the Ly- α forest. c To calculate the DLA number density we counted the z = . z = . d Blended with the Si iv λ more extended than the beam size of the sources which shouldnot be the case for the di ff erent beam sizes to play a role in ex-plaining the excess (Porciani et al. 2007). In addition, we divideour sample in three redshift bins and we find that the numberdensity of strong Mg ii systems is larger in the low redshift bins,favoring the idea that current sample of GRB lines of sight can bebiased by gravitational lensing e ff ect. We also estimate the dustextinction associated to the strong GRB Mg ii systems and wefind that it is consistent with what is observed in standard (sub)-DLAs. It therefore seems that the dust-bias explanation has littlegrounds.We tentatively infer the H i column densities of the strongsystems and we note that the number density of DLAs per unitredshift in the UVES sample is probably twice larger than whatis expected from QSO sightlines. As these sytems are expectedto arise from the central part of massive haloes, this further sup-ports the idea of a gravitational lensing amplification bias. Thishypothesis could be also supported by the results recently ob-tained by (Chen et al. 2008). These authors analyzed 7 GRBfields and found the presence of at least one addition object atangular separation from the GRB afterglow position in all thefour fields of GRBs with known intervening strong MgII galax-ies. In contrast, none is seen at this small angular separation infields without known Mg II absorbers.We searched the Lyman- α forest probed by the UVES spec-tra for the presence of damped Lyman- α absorption lines. Wefound four sub-DLAs with log N (H i ) > z = .
3. This is again twice larger than what is expectedin QSOs. However the statistics is poor. It is intriguing that thesesystems are all located in the half redshift range of the Lyman- α forest closest to the GRB. It is therefore not excluded that partof this gas is somehow associated with the GRBs. In that case,ejection velocities of the order of 10 to 25 000 km / s are required. Acknowledgements.
S.D.V. thanks Robert Mochkovitch for suggesting that sheapplies to the Marie Curie EARA-EST program and the IAP for the warm hos-pitality. S.D.V. was supported during the early stage of this project by SFI grant05 / RFP / PHY0041 and the Marie Curie EARA-EST program. We are grateful toP. Noterdaeme and T. Vinci for precious help.
References
Ballester, P., Banse, K., Castro, S., et al. 2006, in Society of Photo-OpticalInstrumentation Engineers (SPIE) Conference Series, Vol. 6270Barth, A. J., Sari, R., Cohen, M. H., et al. 2003, ApJ, 584, L47Berger, E., Penprase, B. E., Cenko, S. B., et al. 2006, ApJ, 642, 979Bohlin, R. C., Savage, B. D., & Drake, J. F. 1978, ApJ, 224, 132Castro, S., Galama, T. J., Harrison, F. A., et al. 2003, ApJ, 586, 128Chen, H.-W., Perley, D. A., Pollack, L. K., et al. 2008, ArXiv e-prints, 0809.2608Chen, H.-W., Prochaska, J. X., Bloom, J. S., & Thompson, I. B. 2005, ApJ, 634,L25Cucchiara, A., Jones, T., Charlton, J. C., et al. 2008, ArXiv e-prints, 0811.1382Dekker, H., D’Odorico, S., Kaufer, A., Delabre, B., & Kotzlowski, H. 2000, inPresented at the Society of Photo-Optical Instrumentation Engineers (SPIE)Conference, Vol. 4008, Proc. SPIE Vol. 4008, p. 534-545, Optical and IRTelescope Instrumentation and Detectors, Masanori Iye; Alan F. Moorwood;Eds., ed. M. Iye & A. F. Moorwood, 534–545Ellison, S. L., Churchill, C. W., Rix, S. A., & Pettini, M. 2004, ApJ, 615, 118Ellison, S. L., Murphy, M. T., & Dessauges-Zavadsky, M. 2009, MNRAS, 392,998Ellison, S. L., Vreeswijk, P., Ledoux, C., et al. 2006, MNRAS, 372, L38Frank, S., Bentz, M. C., Stanek, K. Z., et al. 2007, Ap&SS, 312, 325Gehrels, N., Chincarini, G., Giommi, P., et al. 2004, ApJ, 611, 1005Greiner, J., Kruehler, T., Fynbo, J. P. U., et al. 2008, ArXiv e-prints, 0810.2314Grevesse, N., Asplund, M., & Sauval, A. J. 2007, Space Science Reviews, 130,105Haislip, J. B., Nysewander, M. C., Reichart, D. E., et al. 2006, Nature, 440, 181Hao, H., Stanek, K. Z., Dobrzycki, A., et al. 2007, ApJ, 659, L99Jakobsson, P., Hjorth, J., Fynbo, J. P. U., et al. 2004, ApJ, 617, L21Kann, D. A., Klose, S., & Zeh, A. 2006, ApJ, 641, 993Kann, D. A., Klose, S., Zhang, B., et al. 2007, ArXiv e-prints, 0712.2186Kawai, N., Kosugi, G., Aoki, K., et al. 2006, Nature, 440, 184Klose, S., Greiner, J., Rau, A., et al. 2004, AJ, 128, 1942Ledoux, C., Petitjean, P., Fynbo, J. P. U., Møller, P., & Srianand, R. 2006, A&A,457, 71Maiorano, E., Masetti, N., Palazzi, E., et al. 2006, A&A, 455, 423Masetti, N., Palazzi, E., Pian, E., et al. 2003, A&A, 404, 465Meiring, J. D., Kulkarni, V. P., Khare, P., et al. 2006, MNRAS, 370, 43Mirabal, N., Halpern, J. P., Kulkarni, S. R., et al. 2002, ApJ, 578, 818Nardini, M., Ghisellini, G., Ghirlanda, G., et al. 2006, A&A, 451, 821Nestor, D. B., Turnshek, D. A., & Rao, S. M. 2005, ApJ, 628, 637Perley, D. A., Li, W., Chornock, R., et al. 2008, ArXiv e-prints, 805P´eroux, C., Dessauges-Zavadsky, M., D’Odorico, S., Kim, T.-S., & McMahon,R. G. 2003, MNRAS, 345, 480P´eroux, C., Dessauges-Zavadsky, M., D’Odorico, S., Sun Kim, T., & McMahon,R. G. 2005, MNRAS, 363, 479Piranomonte, S., Ward, P. A., Fiore, F., et al. 2008, A&A, 492, 775Pontzen, A. & Pettini, M. 2008, ArXiv e-prints: 0810.32362 Susanna D. Vergani et al.: MgII absorbers along GRB lines of sight observed with UVES