Metals in the outskirts of high-redshift galaxies Valentina D - PowerPoint PPT Presentation

Metals in the outskirts of high-redshift galaxies � � Valentina D ’ Odorico INAF – Trieste Astronomical Observatory In collaboration with: G. Becker, F. Calura, R. Carswell, M. Centurion, S. Cristiani, G. Cupani, C. Mongardi, S. Perrotta, E. Pomante, M. Viel, et al. “What Matter(s) around Galaxies” July,19-23 2017

Question driven outline v What are the physical and chemical properties of the CGM? o Metal abundance and distribution o Ionization state v How does the CGM evolve and what can we learn by comparing different epochs? o Redshift evolution of CDDF and Ω v What are the physical processes that shape the CGM? o Feedback mechanisms o Epoch and sources of enrichment and ionization

What is the origin of the observed metals? Enrichment scenarii z~10 EARLY ENRICHMENT v Old metals from previous generations of galaxies è sprinkled in the IGM to low densities, metallicity floor at Z~10 -3 Zo v Fresh metals expelled from coeval galaxies è clustered in the CGM z~2-3 LATE z~2-3 ENRICHMENT

The UVES deep spectrum D’Odorico et al. 2016 Column density distribution functions − 10 Number of lines per unit column This work Ellison et al. 2000 − 11 density and per unit absorption path D’Odorico et al. 2010 − 12 − 13 log dn/dNdX 1 . 0 − 14 − 15 0 . 8 − 16 C IV 0 . 6 ∆ z/ ∆ z tot − 17 0 . 4 11 . 0 11 . 5 12 . 0 12 . 5 13 . 0 13 . 5 14 . 0 14 . 5 15 . 0 15 . 5 log N(CIV) 0 . 2 QSO at z em ~3.0 with V=16.9 0 . 0 T exp =64 h SNR ~300-600 11 . 0 11 . 1 11 . 2 11 . 3 11 . 4 log N(CIV) CIV completeness limits

The UVES deep spectrum D’Odorico et al. 2016 CIV detection rates and connection with galaxies 1 . 0 100 % Metals are not found only in the CGM 0 . 8 (<300 pkpc) of bright star-forming CIV detection rate galaxies at z~2-3 (LBGs): 0 . 6 43 % - They could be present also at larger distances (being produced at 0 . 4 larger z); - They could lie also around smaller 0 . 2 galaxies. 0 . 0 12 . 5 13 . 0 13 . 5 14 . 0 14 . 5 15 . 0 15 . 5 16 . 0 16 . 5 17 . 0 log N(HI) C IV detection rate = # of C IV-H I absorber pairs / # of H I absorbers Fraction of Lyman- α lines in the CGM of LBGs at 2 ≤ z ≤ 2.7, (KBSS, Rudie et al. 2012)

The UVES deep spectrum The metallicity of the IGM 50 12 Cloudy models Z = − 4 . 0 14 Z = − 3 . 0 N CIV 3 Z = − 2 . 0 HM background Z = − 1 . 0 13 log N CIV vs Solar relative abundances (1+ δ )=1 12 N HI T=10 4 (1+ δ ) 0.5 z=2.8 11 14 Assumption: the Jeans N OVI scale is the characteristic vs 13 log N OV I scale of the IGM. Used to N HI 12 transform N HI into (1+ δ ) 11 13 . 0 13 . 5 14 . 0 14 . 5 15 . 0 15 . 5 13.5 14.0 14.8 log N HI Enriched volume to log Z/Zo ≥ − 3: 14.0 ≤ log N HI <14.8 è 2.8 % extending the same Z distribution to 13.5 ≤ log N HI <14.0 è TOT 10.4 % Max volume from OVI è MAX 12.6 %

Mongardi et al. Where are the metals? arXiv:1706.06123 CGM of 20 isolated galaxies in two simulated boxes of 25 comoving Mpc with different prescriptions for star formation and feedback. • Galaxies with M~10 11 -10 12 Mo at z~2 • Boxes pierced with 4000 lines of sight per galaxy with b< 800 kpc • Spectra of HI, CIV, SiIV, OVI built and fitted with VPFIT

Mongardi et al. Where are the metals? arXiv:1706.06123 N(CIV) vs N(HI) N(SiIV) vs N(HI) 12 b < 1 Rvir 1 < b < 3 Rvir b < 1 Rvir 14 1 < b < 3 Rvir 3 < b < 5 Rvir b > 5 Rvir

How does the CGM evolve with time?

CIV cosmic mass density approximated as: Absorption path length interval: Evolution of Ω CIV = metal enrichment X ionisation state of metal-enriched gas C IV becomes a preferred ionisation state at lower overdensities towards higher redshifts (e.g. Oppenheimer & Davé 2006) Shull et al. 2014

CIV cosmic mass density Extension to z~7 by Bosman, Becker et al. (2017) approximated as: Absorption path length interval: Evolution of Ω CIV = metal enrichment X ionisation state of metal-enriched gas C IV becomes a preferred ionisation state at lower overdensities towards higher redshifts (e.g. Oppenheimer & Davé 2006) Shull et al. 2014

Comparison with simulations Finlator et al. (2015): spatially resolved UVB (galaxies + AGNs) Small volume è shortage of strong CIV systems Observational point from D’Odorico et al. 2013 Overproduction of CIV attributed to the shape of the adopted UVB at the HeII edge.

Comparison with simulations Rahmati et al. (2016): EAGLE simulations, homogeneous HM01 UVB Keating et al. (2015): comparison of different simulations/ feedback implementations

SiIV abundance and redshift evolution SiIV could be a simpler element to simulate, weak dependence on UVB shape SiIIV Rahmati et al. 2016 SiIII Adapted from Keating et al. 2015

D’Odorico+ SiIV abundance and redshift evolution in prep. Three samples: v UVES/HIRES 27 QSOs 1.74 < z < 3.7 Δ X~40 (D’Odorico+10, Calura+12) v XQ-100 100 QSOs (45 analysed) 3.2 < z < 4.5 Δ X=99 (Perrotta+16) v Xshooter 7 QSOs 4.95 < z < 6.19 Δ X=27 (D’Odorico+13 plus 1 obj) No of spectra per redshift bin ( Δ z=0.2) Absorption path length Δ X XQ-100 45 obj. 30 UVES/HIRES 27 obj. 30 60 Xshooter 7 obj. Boksenberg & Sargent 2015 25 50 50 Number of spectra 20 40 15 ∆ X 30 10 10 20 20 5 10 0 2 . 0 2 . 5 3 . 0 3 . 5 4 . 0 4 . 5 5 . 0 5 . 5 6 . 0 Redshift 0 2 3 4 5 6 Redshift

D’Odorico+ SiIV abundance and redshift evolution in prep. Three samples: v UVES/HIRES 27 QSOs 1.74 < z < 3.7 Δ X~40 v XQ-100 100 QSOs (45 analysed) 3.2 < z < 4.5 Δ X=99 v Xshooter 7 QSOs 4.95 < z < 6.19 Δ X=27 Column density distribution function − 11 XQ-100 3 . 2 < z < 3 . 8 XQ-100 3 . 8 < z < 4 . 5 − 12 Xshooter 4 . 95 < z < 6 . 19 − 13 log dn/dNdX − 14 − 15 − 16 − 17 12 . 0 12 . 5 13 . 0 13 . 5 14 . 0 14 . 5 15 . 0 log N(SiIV)

D’Odorico+ SiIV abundance and redshift evolution in prep. SiIV cosmic mass density (log N(SiIV) > 12.5) Contribution of DLAs Shull et al. (2014) should be included or Cooksey et al. (2011) Boksenberg & Sargent (2015) not? 10 1 Xshooter 7 obj. XQ100 45 obj. log Ω (SiIV) (x 10 − 8 ) UVES/HIRES 27 obj. 10 1 10 0 No sub/DLAs log Ω (SiIV) (x 10 − 8 ) 10 0 10 − 1 0 1 2 3 4 5 6 Redshift 10 − 1 0 1 2 3 4 5 6 Redshift

Redshift evolution of CGM metals The mass density of CIV evolves significantly towards lower redshift while SiIV seems to have a jump at high redshift. Effect of different environment or UVB, or…? Log N (CIV) > 13.4 Log N (SiIV) > 12.5 10 1 Xshooter 7 obj. XQ100 45 obj. UVES/HIRES 27 obj. log Ω (SiIV) (x 10 − 8 ) 10 0 10 − 1 1 2 3 4 5 6 7 Redshift

Final thoughts & future perspectives ü Metals are not confined to the CGM of bright star forming galaxies: should we change the definition of CGM or maybe define a metallicity threshold between CGM and IGM? ü Data suggest pre-enrichment; ü Present simulations have problems in reproducing the redshift evolution of Ω CIV and the properties of CIV at z~6; ü SiIV could be a simpler element to simulate, its behaviour with time is different wrt CIV. Medium/High-resolution spectroscopy with 8-10m class telescopes has reached the “photon starving” regime for many of the IGM hot topics è z~2-3 ESPRESSO@VLT online in 2018 z~6 and beyond HIRES@ELT in 2025-2030