The Big Three
Virgo, Coma, and Perseus from UHURU to Chandra
Bill Forman (SAO-CfA)
•
•
•
•
•
Brief reminiscence from UHURU
Perseus/Coma/M87 from Chandra
M87 - ROSAT, XMM, Chandra
M87 from Chandra with Jones/Churazov/Heinz
•
•
•
•
Outburst up close
Classic shock
Buoyant bubbles
Collaborators:
Christine Jones, Eugene Churazov,
Sebastian Heinz, Ralph Kraft, Akos Bogdan, Mike Anderson,
Paul Nulsen, Scott Randall, Larry David, Jan Vrtilek, Simona
Giacintucci, Marie Machacek, Ming Sun, Maxim Markevitch,
Alexey Vikhlinin
Energy partition and outburst duration
Early type galaxies with SMBH (Jones, Churazov, Anderson)
•
•
•
Feedback present in X-ray/optically luminous galaxies
Hot X-ray coronae - mechanism to capture SMBH energy
Driver of galaxy evolution
1973ApJ...184L.105L
Coma from Uhuru
(Forman et al. 1972)
Lea et al. 1973
• Clearly extended
• Nature unknown
• Lea et al. 1973 “There is no
compelling evidence that
the emission is due to
thermal bremsstrahlung
from a hot gas ….”
2
Perseus from Uhuru
(Forman et al. 1972)
• Clearly extended
1972ApJ...178..309F
• Nature unknown
Ariel 5
• Mitchell, Charles, Culhane, Davison, Ives et
al. 1976, MNRAS, 176, 29
• Emission suggests 6 keV plasma enriched
with iron
1972ApJ...178..309F
• Quickly led to “cooling flow problem” e.g.,
Fabian & Nulsen 1977
Off we go
- and we
never
looked
back!
3
Coma - mini-coronae (Vikhlinin+01)
•Mini
NGC4889
NGC4889
Chandra
NGC4874
coronae around BOTH
central cDs (Vikhlinin+01)
•3 kpc radius; 108 Msun; ~1 keV
•pressure confined; thermal
conduction suppressed by
factor of 30-100
• Sun+07 mini-corona survey
•25 hot (>3keV) clusters
•60% of >2L* galaxies (~100)
have mini-coronae
•X-ray fainter than “typical”
coronae
NGC4874
Chandra
Coma (XMM) - merging
Coma (ROSAT) - merging
Tail - to NGC4911
Vikhlinin+00
NGC4889
NGC4874
NGC4839
0.5 Mpc
•Mgas~5x1011 Msun; kT ~ kTComa
•Origin
•Ram pressure stripped gas?
•Cluster gas compressed in tidally
stripped dark matter filament?
NGC4839
•Possible slightly supersonic merger
•Suggested by hot X-ray sheath
(kT~6.2-6.7 keV; kTambient~4.8 keV
•∆v~1700 km/s (Colless+07) ==> M~1.3
•400 kpc long tail
•Mgroup >1014 Msun
Virgo Cluster and M87 - the main course!!
H. Boehringer
M87 is central dominant galaxy
1’=4.65 kpc; 2o=0.5 Mpc
Optically luminous early-type galaxies
are (hot) gas rich - up to 1010 Msun
Virgo is dynamically young
extensive merging, stripping
•Clear from X-ray image
•M87 is 50 x more X-ray luminous than
NGC4472
•NGC4472 (a bit) optically more luminous
than M87 aka don’t believe everything you
“see” (optically)
•M87 hosts 6x109Msun supermassive black
hole and jet
•Classic cooling flow (24 Msun/yr)
•Ideal system to study SMBH/gas
interaction
SLOSHING IN M87
Gas Sloshing in M87 (XMM)
THE WINE
courtesy
of W. Forman
M87 shows
gas “sloshing”
“Edge”, contact discontinuity - cold front at ~100kpc
(Simionescu+10 from XMM-Newton)
Einstein Fellows Symposium
CfA, Oct. 28, 2009
Aurora Simionescu
Very common (14/18) in “peaked” clusters (Markevitch+03)
see Markevitch & Vikhlinin 2007 for a review
Driven by mergers
Markevitch&Vikhlinin 2007
Gas Sloshing in M87 (ROSAT)
THE WINE
Sloshing edge/contact discontinuity visible in ROSAT!
Note - after the fact, you can see the Bullet cluster/bullet
as well.
Markevitch&Vikhlinin 2007
X-ray and Radio View of M87
•
Multiple - at least three - AGN outbursts
•
Two X-ray “arms” - produced by buoyant radio
bubbles
• Eastern arm - classic buoyant bubble with
torus i.e., “mushroom cloud” (Churazov et al
2001)
– XMM-Newton shows cool arms of uplifted gas
(Belsole et al 2001; Molendi 2002)
M87
Forman+05,+07
Million+10, Werner+10
X-ray Temperature Map
Radio 90Mhz
Owen, Eilek, Kassim 2001
X-ray and Radio View of M87
• At least four major SMBH outburst events
•
Large radio “bubbles”
• Radio/X-ray “arms” - produced/uplifted by
buoyant radio bubbles
• classic buoyant bubble with torus i.e.,
“smoke ring” (Churazov et al 2001)
• Shock at 12 kpc - initial inflation of bubble
• Current/ongoing outburst
•Radio 90Mhz
Old bubbles
•Owen et al. 2001
•Owen+00 VLA
•Forman+05,+07
•Million+10, Werner+10
Fate of Bubble Energy
Rising bubble loses energy to
surrounding gas
Generates gas motions in wake
Kinetic energy (eventually)
converted to thermal energy (via
turbulence)
non-relativistic
Bubble energy
remaining
vs. radius
ΔEgas = −ΔEBubble
relativistic
1−1 / γ
&
#
, P)
γ
!
= −Δ
PV = E0 $1 − ** ''
γ −1
$% + P0 (
!"
Classical Shock in M87
2
∫ P dl
Xarithmetic (Churazov et
al. 2015)- choosing proper
band
Piston drives shocks
SHOCK
Chandra (0.5-1.0 keV)
Chandra (3.5-7.5 keV)
23 kpc (75 lyr)
• Black hole = 6.6x109 solar masses (Gebhardt+11)
• SMBH drives jets and shocks
• Inflates “bubbles” of relativistic plasma
• Many small bubbles
• Heat surrounding gas
• Model to derive detailed shock properties
Central Region of M87 - the driving force
SMBH 3x109Msun
6cm radio
“Bud”
• Cavities surround the jet and (unseen) counterjet
• Bubble breaking from counter jet cavity
– Perpendicular to jet axis;
– Radius ~1kpc.
– Formation time ~4 x106 years
• Piston driving shock
– X-ray rim is low entropy gas uplifted/displaced by
relativistic plasma
6 cm
Shock Model - the data
•Hard (3.5-7.5 keV) pressure
•
soft (1.2-2.5 keV) density profiles
• Projected
•Deprojected
Gas Pressure (3.5-7.5 keV)
14
Textbook Example of Shocks
Consistent density and temperature jumps
Rankine-Hugoniot Shock Jump Conditions
γ + 1) M 2
(
ρ 2 / ρ1 =
(γ + 1) + (γ −1)( M 2 −1)
ρ 2 / ρ1 = 1.34
€
(γ + 1) + 2γ ( M
[
T /T =
2
1
€
2
−1)
][(γ + 1) + (γ −1)( M
(γ + 1)
2
2
−1)
]
M2
T2/T1= 1.18
yield same Mach number:
(MT=1.24 Μρ=1.18)
M=1.2
Outburst Model - grid in total energy and duration
Forman et al. 2015
Etot = 1.4, 5.5, 22x1057 ergs)
duration = 2.2 Myr
Match all constraints
Etot = 5.5x1057 ergs, durations
= 0.1, 1.1, 2.2, 3.1, 4.0, 4,4, 6.2
Myrs
Shock strength (nearly)
governed by Etot
Characterizing M87’s outburst Long vs. Short Durations
Forman et al. 2015
0.6 vs 2.2 Myr duration outbursts with
Eoutburst = 5.5x1057 ergs
Short outburst - leaves hot, shocked
envelope outside the piston
NOT observed
longer duration
outburst required
Rapid Piston
(Relativistic Plasma)
Outer
corona
Piston
Slow Piston
(Relativistic Plasma)
Outer
corona
Shock
Shock
Shock
Strongly
Shocked Gas
Weakly Shocked
Gas
Outburst Model - 3D verification (as in Heinz+06, Morsony+10)
s
−of−
line
ight
0.010
Density, cm−3
0.10
density, cm−3
cut plotted in Fig. 14
0.100
0.01
0.001
0
1
• Full 3D model - view at 200
• 1044 erg/s jet - now
Temperature, keV
4
3
2
1
0
1
10
• 3D produces elliptical cavities
• Total energy increased by 50%
• After jet turns off, core refilled with slightly hotter plasma
• Excellent match to T(r), n(r) profiles
Radius, kpc
M87 Outburst Energy Parameters
Detect shock (X-ray) and driving piston (radio) (Forman et al. 2007)
Classical (textbook) shock M=1.2 (temperature and density
independently)
Outburst constrained by:
Size of driving piston (radius of cocoon)
Measured T2/T1, ρ2 /ρ1 (p2/p1)
Current shock radius
Current cavity size
Outburst Model (Forman et al. 2015)
Age ~ 12 Myr
Energy ~ 5x1057 erg
Bubble 50%
Shocked gas 25% (25% carried away by weak wave)
Outburst duration ~ 2 Myr
Outburst is not violent (not Sedov-like)
Outburst energy "balances" cooling (few 1043 erg/sec)
Feedback from Supermassive Black Holes
key component in galaxy formation models
% of gals that are radio-loud AGN
100
SN feedback+photoionization
10
Dark halos
(const M/L)
galaxies
AGN feedback
LNVSS > 1023 W Hz-1
LNVSS > 1024 W Hz-1
LNVSS > 1025 W Hz-1
1
0.1
0.01
0.001
9.0
9.5
10.0
10.5
11.0
11.5
Stellar mass (solar masses)
12.0
100
% of gals that are optical AGN
• Feedback - mass closely tied to mass of surrounding stars - MSMBH ≈
10
10-3Mbulge
• SMBH key to regulating star formation in evolutionary
models at high
1
mass end
• Radio loud AGN very common in massive galaxies -0.1“radio mode” vs. L > 10
L
> 10
L
> 10
quasar mode (Churazov+05)
0.01
[OIII]
[OIII]
[OIII]
5.5
Lsun
Lsun
7.5
Lsun
6.5
9.0 Best+06,
9.5
10.0 Teyssier+11
10.5
11.0
11.5
e.g. Croton+06, Bower+06,White & Frenk 91, Cole+92 Benson+’03
Stellar mass (solar masses)
12.0
Figure 2. Top: the fraction of galaxies which are radio–loud AGN, as a funct
Early type galaxy sample from Jones et al. (Anderson, Churazov, Forman+)
LX/LK vs. LK
NGC1316 = Fornax A
NGC4291
---------
------
•Fomalont/NRAO
NGC4342
350 kpc
•Lnuc~2x1042 erg/s
•Massive SMBH is
• Early type galaxies are gas rich
• Cavities common > 30% in luminous systems
• SMBH detected 70% radio and 80% X-ray
• Winds at LK < 1011
• Low Eddington ratios ~10-5 - 10-9 in these low
luminosity AGN
(for QSO’s ~0.3) (Eddington ratio for Sag A = 10-9)
willing and
able to disrupt atmosphere given
sufficient fuel; outburst power ~
5x1058 ergs (Lanz+10)
•Gas
rich mergers could drive
such outbursts at early epochs
and disrupt star formation
Optically faint, gas rich galaxies - NGC4342
ROSAT PSPC Image
(Boehringer & Schindler)
NGC4342 beyond r200 from M87
Only ~0.5 Mpc from NGC4472 (M49)
Virgo gas distribution - elongated N-S
Gaseous filament in Virgo outskirts
NGC4342 encounters external gas for
the first time?
Ram pressure stripping underway?
ÁN ET AL.
Massive Black Holes (Bogdan et BOGD
al. 2012)
- two outliers
4
9
10
NGC4342
8
10
M
MBHSMBH
, Msun
NGC4291
o
7
la
e
n r
7
10
n
a
n
s
i
d
d ea
6
10
9
10
M
s
r
pe
•
i on
•
Fornax A at early epochs
where Tgas all
and stars
ne are formed
radial profil
BEFORE
NGC4342
NGC4291
10
10
11
Mbulge, Msun
M
10
The 0.5 − 2 keV band X-ray image of
a diffuse hot gas component associated
Evolutionary
scenario for
which exhibits a significantly
broader
the stellar light
(Fig.
1). To compu
NGC4342
and
NGC4291
mass profile of NGC4342 we assume t
in hydrostatic
equlibrium (Mathews 19
Star
formation
1985; Humphrey et al. 2006) and use t
tion:
suppressed
by powerful
!
∂ ln ne
kTgas
(r)r like
SMBH
outburst
(e.g.,
Mtot (< r) = −
Gµmp
∂ ln r
12
10
Fig. 3.— Black hole mass as aBULGE
function of bulge mass. Thick
solid line shows the mean M• − Mbulge relation from Häring & Rix
(2004), whereasand
the thin
dashed linehost
represent
the intrinsic
scatter
•NGC4342
NGC4291
massive
dark
of the relation. Both NGC4342 and NGC4291 are highly significant
outliers from
the trend.
matter
halos
sufficient to bind hot coronae
•measured
viaarehydrostatic
black
hole masses
7.7 × 106 M⊙equilibrium
and 7.4 × 107 M⊙
NGC4342
andare
NGC4291,
respectively.
• in
Black
holes
too massive
for Thus,
theirthe observed values are factors of ≈60 and ≈13 times larger
bulges
and ones.
13x larger
than the(60x
predicted
From thethan
intrinsic scatter of
the relation (0.30 dex) and the uncertainty of the black
“predicted”)
hole mass measurements (0.18 dex and 0.12 dex), we conclude that NGC4342 and NGC4291 are ≈5.1σ and ≈3.4σ
•
•
temperature and density, respectively
projected growth
profiles we
use the techni
SMBH
precedes
Churazov et al. (2003). Namely, we m
spectra ascomponent
the linear combination
stellar
e.g., of sp
shells plus the contribution of the outer
Sijacki+14
that emissivity in the outer shell declin
with radius at all energies. The mat
CID
947 (Trakhtenbrot
the projection
of the shells into annu
the deprojected spectra are calculated
+15)
possibly
inverted
matrix to the observed spectr
Due the head-tail distribution of th
inventory
tingeRosita
gas (Fig. will
1 right
panel; Bogdán
assumption of hydrostatic equilibrium
dark matter halos
radii larger than ∼5 kpc. To accoun
tainties we computed mass profiles in t
tors: (i) towards the surface brightness
• M87 classic shock and bubbles
– reveals detailed SMBH interaction
Review
M87 - bubbles & shocks
– shocks are typically “weak”
X-ray (soft & hard)
– outbursts are “long” (>Myr)
– bubbles carry most of energy (>50%)
• AGN outbursts are common in all gas rich systems
• bubbles/cavities everywhere!
• more massive systems are more likely radio
bright
NGC4342
• “cooling flows” from galaxies (~1 Msun/yr) to clusters
(~few 100 Msun/yr) moderated by SMBH energy
release
• SMBH’s are willing and able to disrupt cooling
atmospheres at low (and possibly high) redshifts
(NGC4342/NGC4391 SMBH’s are too massive for
their stellar mass)
• SMBH outbursts are a key phenomenon across a
vast range of halo mass and cosmic time
galaxies
groups
clusters
Mhalo ~ 1012 —> 1015 Msun
Supermassive Black Hole Outbursts in the Family of
Early Type Galaxy Atmospheres
Jones+
Galaxy
1 kpc
1056 ergs
1042 erg/s
McNamara+
Fabian+
Group/Cluster Core
10 kpc
1059 ergs
1045 erg/s
Cluster (MS0735)
100 kpc
1062 ergs
1046 erg/s
Very powerful outflows
Very little radiation from black hole
Gas cooling rates vary by > 100x
Span a wide range of dark matter halo mass
Zhuravleva+14 - Solving the “cooling flow” problem
•
•
•
•
tcool is << tage
More than enough energy from SMBH in buoyant bubbles & shocks
Plus mergers and gas sloshing
But how, exactly, does the energy transfer occur?
• Measure power spectrum of surface
brightness fluctuations
• Deproject to get density fluctuations
• 1D gas velocity
∝ rms density
fluctuations
• Turbulent heating is sufficient to
offset radiative cooling
• Balances locally at each radius
• May be key to heating hot coronae
from clusters to early type galaxies
Figure 3 | Turbulent heating (Qheat) versus gas coo
cores. Each shaded rectangle shows the heating and
(top right – the innermost radius; bottom left – the o
Finis
as we look forward to the next generation of
X-ray missions and discovery
10
Churazov et al.
28