Martin Haehnelt Probing reionization with Lyman-alpha and 21cm emission/absorption
MartinHaehnelt
Probingreionizationwith
Lyman-alphaand21cmemission/absorption
DominiqueAubertGeorgeBeckerJamesBoltonJonathanChardinTirthChoudhuryHarleyKatzLauraKeatingTaysunKimmGirishKulkarniEwaldPuchweinAlbertoRorai
ProbingtheEpochofReionization
Ø CMBØ Lyman-alphaabsorptionØ Lyman-alphaemitters/highredshiftgalaxiesØ 21cmemission/absorption
Robertsonetal.
Hominginonthereionizationhistory
Robertsonetal.2013,Ellis2014 Chornocketal.2014
1-QHII fHI
Buthowearlydoesitstart,howlatedoesitfinish?
Beijing, 30 August 2016
High-redshiftgalaxies
High-redshiftgalaxies
Bouwensetal.2014
Ø Dramaticprogresswithcharacterisationofluminosityfunctionofcontinuum-selectedgalaxiesatz>6Ø NoideawhatescapefractionofionizingphotonsØ WithJWSThopefully/possiblyescapefractionsfromHαvsLyα
escapefraction?
Beijing, 30 August 2016
The“standard”UVbackgroundmodel
ionizingemissivity
galaxiesQSOs
Thismayneedsomemodification.
Haardt&Madau2012–Thebenchmarkmodel
Beijing, 30 August 2016
CMB
ThePlanck2015/16resultsforτ
Ø constraintsfromCMBhavebecomeratherweak.Ø jointconstraintwithBAOappearstocomedown
Planck2016àverylateandfastreionization?
Beijing, 30 August 2016
Lyman-αemissionfromhigh-redshiftgalaxies
Stark2011
Konnoetal.2014
Beijing, 30 August 2016
Lyα absorption by neutral hydrogen
λobs = 1216 (1+z) Å
Hydrogen in the IGM is photoionized:
α nHII ne = Γ nHI
Recombination Photoionization
Dijkstra2014
TheLyman-alphascatteringcrosssections
Thereareprominentdampingwings!
DampedLyman-alphaabsorption
Barnes&Haehnelt2010
Resonantscattering
Barnesetal.2010
strongwinds weakwinds
Barnesetal.2011
TheLyman-alphaemissiondependsstronglyonthestrengthofagalacticwind.
weakwind
strongwind
Probingoutflows/inflows
Dijkstraetal.2011
TheshapeofthespectralprofileandtheredshiftoftheLyαemissionescapingfromthehostgalaxyisveryuncertain.
ionizedifζefffcoll(x,R)>1
hybridsimulation:largeDMboxplusSmallhigh-resolution
TopredictLyαtransmissionisdifficult!
z=7
Choudhury,Puchwein,Haehnelt&Bolton2015
Choudhury,Puchwein,Haehnelt&Bolton2015
Ø TherapiddeclineofLyαemissionatatz>6requiresa(significant)declineofthevolume-weightedionizedfractionperhapsaidedbyanevolutionoftheintrinsicredoffset
Ø Reionizationappearstofinish(somewhat)laterthanpredictedbyHM2012Ø ShouldbepossibletoprobebubblesizeandtopologyofreionizationwithHyperSuprimeCam
TherapiddemiseofLyαemissionatz>6
Beijing, 30 August 2016
Ly-αabsorption
z=6.42
z=5.09
z=5.20
z=5.42
z=5.80
z=5.82
z=6.30
Fanetal.,Beckeretal.
τGP ≈ 105 nHI
nH
⎛
⎝⎜
⎞
⎠⎟
SearchingforaGunn-PetersontroughinQSOabsorptionspectra
Boltonetal.2011
Near-zonesizes
HIIregionorclassicalproximityeffect?
€
R∝ ˙ N tQ f HI−1( )
1/ 3
Veryhardtotellthedifference!
ExpandingintoneutralIGM
€
R∝ ˙ N 1/ 2
ExpandingintohighlyionizedIGM
z=7.085
Mortlocketal.2011
Areddampingwing?
ZoomsbasedonMilleniumsimulation
average intermediate mostmassivehalo
z=7
10h-1Mpc
Boltonetal.2011
Near-zonesizes
Theobservedtransmissionprofileappearsconsistentwitha10%neutralaswellasafullyneutralsurroundingIGMdependingonquasarage.
Theobservedtransmissionprofile
observed
106yr
observed
107yr
Boltonetal.2011
Edgeofnear-zonesaredefinedbylocationofsuperLLS
Keating,Haehnelt,Cantalupo&Puchwein2015
Nicelyexplainstheobservedscatterinnear-zonesizes!
Mockspectraofnear-zoneregions
Keating,Haehnelt,Cantalupo&Puchwein2015
DegeneracybetweenneutralfractionandQSOlifetime
Near-zonesizesatz=7
Keating,Haehnelt,Cantalupo&Puchwein2015
Near-zonesizeofULASJ1120+0641isanomalouslylowunlesstheQSOhasbeenbrightforonlyarathershorttime.
ULAS1120+0641
Formassivehosthaloesthereddampingwingisnotparticularlysignificant.
Thereddampingwing
ULAS1120+0641
CarefullycalibrattedwithLyαforestdata
RAMSESpost-processedwithATON
Photo-ionizationratewellreproducedwithsomewhatmodifiedionizingemissivitycomparedtoHM2012.Reionizationoccurssomewhatlater.
ionizingemissivity
photo-ionizationrate
Chardinetal.15
ReionizationappearstooccurlaterthaninHM2012
Chardinetal.
SherwoodSimulationSuiteBoltonetal.2016
2x8.5billionparticles5runstoz=2,40-160Mpc/hboxes15,000,000core-hoursfromPRACE
Haardt&Madau2012
Late(Chardinetal.15,16)
VeryLate(Planck2016?)
Kulkarnietal.2016,submitted
Beijing, 30 August 2016
21cmemission
Measurementsaregettingclose!Kulkarnietal.2016,submitted
IncludingtheHIinself-shieldedregionsinionizedbubblesmakesadifference.
Kulkarnietal.2016submitted
Beijing, 30 August 2016
LargescaleLyαopacityfluctuations
Beckeretal.2015
slidefromJonathanChardin
Therearelargefluctuationsoftheopticaldepthonsurprisinglylargescales.Aretherestilllargecompletelyneutralregionsevenatz~5.5?
Beckeretal.2015
ionizingemissivity
galaxiesQSOs
TheremaybemoreQSOsathighredshiftthanassumed.
Giallongoetal.2015
Haardt&Madau2012–Thebenchmarkmodel
12
Chardinetal.162016arXiv160608231
OpacityfluctuationonlargescalaesàQSOs?
Chardinetal.162016arXiv160608231
IsthebrightendofthegalaxyluminosityfunctiondominatedbyAGN?
12
Chardinetal.162016arXiv160608231
OpacityfluctuationonlargescalaesàQSOs?
Droppinggalaxiesalltogetherisperhapsabitradical.SoftX-raybackgroundprobablyOK.Implicationsforheliumreionizationstilltobeexplored.
Summary
• evidenceisbuildingforratherlatereionization
• broadfluxPDFatz≈5.6-5.8mayrequiresubstantialcontributionofQSOstophotoionizationrate
• perhapssomemoreattentionofobserversshouldbeshiftedtodetermineΦ*(QSO)andtolookforthetransverseproximityeffectatz≥5• 21cmsensitivityat6<z<10isgettingclosetomodelpredictions