The Lyman- halo of B2 0902+34: Evidence for infall of extended HI Joshua Adams & Gary Hill University of Texas, Austin Department of Astronomy 3D2008 ESO Workshop Garching, 6-13-08
Jan 12, 2016
The Lyman- halo of B2 0902+34:
Evidence for infall of extended HI
Joshua Adams & Gary Hill
University of Texas, Austin
Department of Astronomy
3D2008 ESO Workshop
Garching, 6-13-08
Overview
• High Redshift Radio Galaxy (HzRG) Lyman- halo science motivation
• New integral field spectroscopy data on B2 0902+34
• New Monte Carlo resonant scattering model
• Future observations and model discrimination
Why bother with HzRG halos?“While LABs are known to exist around radio-loud quasars, they are attributable to known jets and supernova-driven outflows, and we do not attempt to model them here.”
Dijkstra et al. (2006b)“(The LSBHs) which extends across the entire object and beyond the edge of the radio lobes shows no apparent association with radio structures.”
Villar-Martin et al. (2002)
Massive galaxy and cluster formation/feedback
MRC 0052-241, Venemans et al., 2007 Rawlings and Jarvis, 2004
HzRG Lyman background
• van Ojik et al. (1997) find spatially resolved line profile structure in 11/18 HzRGs
• Villar-Martin et al. (2003) find LSBHs in 10/10 HzRGs + at least 4 more known
• Similarly, 5 quasars (Christensen et al. (2006),Weidinger et al. (2004)) show extended emission
• Unknown relation, if any, to LABs
Existing B2 0902+34 data
Reuland et al. 2007
Exciting new spectral feature
•Inclination of northern radio jet is >30˚ and <45˚ (Carilli 1995)
•21 cm HI absorption at z=3.3968 with FWHM=120 km/s and N=3x1021 cm-2 (Uson et al. 1991 and others)
•Normal line ratios for pure AGN photoionization in Ly-, CIV1549, and HeII1640 (Villar-Martín et al. 2007)
Existing B2 0902+34 data
Reuland et al., 2003
VIRUS-P instrument• Large fibers with 2.7m focal
reducer: 4.2” diameter• 105x105arcsec2 field, fill
factor 1/3, 247 fibers• 3500-5800Å, R~1000
No resampling in reductions
• Does not introduce correlated noise
• Allows sky subtraction free of any features and reaches noise limits
Liner interpolation Bspline, similar to Kelson (2003)
Our VIRUS-P data
a
b
c
Radio data from Carilli 1995
Our VIRUS-P dataPrimary emission:
5339.0 ± 2.0 Å
600 ± 90 km/s FWHM
Secondary emission:
15% as strong
5324.5 ± 1.7 Å
630 ± 270 km/s FWHM
Failing explanations for B2…• Optically Thin Infall
– No bimodal line profile allowed– Too small FWHMs
MRC 1558-003, Villar-Martin et al., 2007
Q1205-30, Weidinger et al., 2004
Failing explanations for B2…
• Outflow (Reuland et al. 2007)– Where is the 21cm HI population in Ly-?– Why is there no line profile bimodality in the
NE? Wilman et al., 2005
Our model: resonant scattering
• Monte Carlo Resonant Scattering code with 7x105 photons per simulation
• Biconal emission geometry per the Alignment Effect (McCarthy 1993)
• Isothermal NFW profile, baseline simulation uses 6x1012 Ms halo with rv=134 kpc and ri=97 kpc
Model Geometry
Tunable parameters: 1) total halo mass 2) ionization radius 3)velocity strength 4)velocity power law with radius
Monte Carlo radiative transfer• Pick a random optical depth from exponential
deviate in random direction• Transform to scattering particle’s rest frame• Obtain scatterer’s parallel velocity from the following
pdf:
• Obtain scatterer’s Maxwellian perpendicular velocity• Obtain scattering direction from dipole distribution• Transform back to observer’s frame• Repeat
du
aux
e
axH
aduuf
u
22
2
,
Resonant scatter trends
Blue Red
0
33
2
0
272
cosh124
a
x
x
axJ
Far
Cone
Near
Cone
Resonant scatter trends
Blue Red
0
33
2
0
272
cosh124
a
x
x
axJ
Far
Cone
Near
Cone
Resonant scatter trends
Blue Red
0
33
2
0
272
cosh124
a
x
x
axJ
Far
Cone
Near
Cone
Our VIRUS-P data + model
a
b
c
Radio data from Carilli 1995
Our VIRUS-P data + model
Surface brightness
Conclusions• A resonant scatter model explains
– The spatial distributions of the bimodal profile– The relative intensities of the bimodal profile– The relative wavelengths of the bimodal profile– The surface brightness profile– The 21cm data
• We predict a very large HI mass, 1011-1012Ms
• VLBI radio spectral imaging can falsify our model• Are B2 0902+34 and other HzRGs displaying
– Strong AGN feedback (outflow) or– Large (cluster?) galaxy formation (infall)
Extras
The 21cm Data,mk picObservation Our Model
N (cm-2) 3x1021 7x1021
FWHM (km/s) 120 273
V (km/s) - 121
A trial system: B2 0902+34
• Carilli (1995) derives a low inclination angle for the radio jets– Hot spot A is bright and
polarized– High projection explains
steep spectral index in north
– High projection explains lack of alignment effect
Systemic redshift estimates
Litmus Test on Infall: Skew
Infall or Outflow
Dijkstra et al. (2006b)Wilman et al. (2005)
General Resonant Scatter Results
Dijkstra et al. (2006a)
• “Dip” Due To Doppler To Line Center Against A Velocity In Neutral Hydrogen
• Velocity Magnitude May Come From Red Bump Position/Relative Magnitude
• Velocity Shape May Come From Wavelength Dependent Surface Brightness Profile
Existing B2 0902+34 data
• Undetected in CO(4-3) and CO(5-4), along with 13 other HzRGs (van Ojik et al., 1997) and in CO(4-3), CO(5-4), and CO(8-7) in Evans et al., 1996
• Spitzer data
(Seymour et al., 2007)
gives fstel=0.28 and
Mstel=1010.81Ms
Code Tests
Emergent Spectra From Static Sphere
Redistribution Function with Dipole Phase Function