A Study of Sgr A and Nonthermal Radio Filaments F' YusefZadeh

1
A Study of Sgr A and Non-thermal Radio
Filaments F. Yusef-Zadeh
2
Outline

• Sgr A
• Light curves in NIR wavelengths
• Sub-mm emission is variable
• Cross correlation
• Power spectrum of NIR emission
• Non-thermal Filamen ts
• Recent observations
• Global vs local origin
• Strong vs weak magnetic field
• a new turbulent model

X-ray
NIR
Sub-mm
3
Two Epochs of Observations of SgrA in 2004
March Campaign
SMT
Sub-mm
CSO
BIMA
mm
NMA
VLA
Radio
ATCA
XMM
X/g-ray
INTEGRAL
September Campaign
CSO
Sub-mm
VLA
Radio
HST
NIR
XMM
X/g-ray
INTEGRAL
4
Near-IR Line and Continuum (NICMOS /HST)

• Observations
• 32 orbits of Camera 1 NICMOS in three bands
• F160W Broad band at 1.6mm (1.4-1.8mum)
• F187N Narrow band at 1.87mm(1.865-1.885)
  (Pa line)
• F190N Narrow band at 1.9mm (continuum)
• Each cycle Alternating between the three bands
  for about 7-8 min
• Six observing time intervals

5

• 1.6 mm image
• Pixel size 0.043
• Field of view 11
• Sigma0.002 mJy after 30sec
• The position of S2 wrt SgrA estimated from
  orbit calculations (Ghez et al. 2003)
• S2 offset from SgrA is 0.13N and 0.03E
• PSF has a four pixel diameter
• S2 is a steady source (PSF contamination)
• Great instrument because of background and PSF
  stability
• PSF has a 4-pixel diameter

6
Near-IR variability in 1.6, 1.87 (Paa line),
1.9mm

• Near-IR Light Curves of SgrA (blue, red, green)
• Amp 10 to 25 or 3 to 5 times the quiescent
  flux (11-14 mJy)
• Duration multiple peaks, lasting from 20
  minutes to hours
• Flare activity overall fraction of activity is
  about 30-40 of the observed time (background 8.9
  mJy)

a
b
c
d
e
7
Submillimeter 0.87mm and 0.45mm Emission (CSO)
September Campaign

• Variability with maximum 4.6Jy and minimum 2.7
  Jy
• Likely to be due to hourly than daily variability

8
IR (1.6-1.9mm) vs. X-Ray (September Campaign)
NIR
X-ray
X-ray
NIR
9
Simultaneous X-ray and NIR flare

• NIR due to Synchrotron tnir 40min, Beq10G,
  Ee1.1 GeV
• X-Rays due to Synchrotron tnir1min, B10G,
  Ee10 GeV
• X-Rays due to ICS diameter10Rsch, F850mm4Jy,
  Ee1GeV
• Ex2x10-12 erg/cm2/s/keV, Eobs1.2x10-12
  erg/cm2/s/keV
• Spectral index between near-IR and X-rays is a
  -1.35 bu in near-IR and X-rays , a -0.6
• If a is flatter by 0.2, the X-ray flux is
  reduced by a factor of 2, so no X-ray
  counterpart is detected.
• ICS with a -0.6 produces soft g-ray emission
  with L 7.6x1034 erg/s
• (2-20 keV) which is 5 times lower than
  observed

10

• Lomb-Scargle periodogram searches for
  periodicity
• Significant power with a period of 1.3h and
  30min
• a/M0
• (r/M)orbit 6Rsch
• The same scale size as the region of the seed
  photons for ICS

11
Radio (7mm) vs X-ray (March Campaign)
X-ray

• Lag time between X-ray/NIR flare and sub-mm peak
  4-5 hours
• Time delay between X-ray/NIR and radio peak is
  one day
• The flux variation in sub-mm is about 30 whereas
  at 7mm is 10
• An expanding synchrotron source in
• an optically thick medium

NIR
Sub-mm
Radio
12
Sub-mm and Radio Time Lags
D.P. Marrone, 2005
13
Conclusions I

• Flare correlation simultaneous vs delayed
• Correlation between a near-IR and X-ray flare
  consistent with SSC (Eckart et al. 2004)
• An expanding synchrotron self-absorbed blob
  radio and NIR wavelengths
• Evidence for a NIR flare with quasi-periodic
  1-1.3h and 30min behavior
• The flow always fluctuates even in its quiescent
  phase

14

• Non-thermal radio continuum emission is
  significant (40-50)
• Excess SNRs
• Stellar clusters (Arches cluster, Sgr B2)
• Diffuse background emission
• Magnetized Filaments

Law et al. (2005)
15
The ripple filament total intensity (top),
polarized intensity at 6cm (bottom)
16
X-ray (2-8 keV) and Radio (15GHz) Images of SgrA-E
X-ray
Radio
Sakano et al. 2003 Lu et al. 2003 Yusef-Zadeh et
al. 2005
17

• The spectrum is similar in both radio and X-ray
• Using the 1200, 862 and 442 micron flux and RJ
  tail of Planck function
• Tdust 30-50 K
• t(dust) 0.008-0.004
• Using 25 mJy flux at 15 GHz with a0.75, ICS flux
  can match the observed X-ray flux 3.2x10-14
  ergs/ cm2/s/keV if
• Tdust 50 K, B100 m G
• Beq 140 m G for a cutoff of 10MeV

18
Distribution of Radio Filaments
19
Origin of the Filaments

• Models
• Loop ejection from a central differentially
  rotating engine at the Galactic center
  (Heyvaerts, Norman and Pudritz 1988)
• Induced electric field by the motion of molecular
  clouds with respect to organized poloidal
  magnetic field (Benford 1988)
• Contraction of a rotating nuclear disk in a
  medium threaded by poloidal magnetic field
• Cosmic Strings oscillating in a magnetized medium
  
• Interaction of a magnetized galactic wind with
  molecular clouds (Shore and LaRosa 1999)
• Reconnection of the magnetic field at the
  interaction site of molecular clouds and
  large-scale magnetic field (Serabyn and Morris
  1994)

20
Magnetic Field in the Galactic Center

• Standard Picture for the last 20 years
• The field has a global, poloidal geometry due to
  the filaments orientation
• The field has a milli-Gauss strength due to
  dynamical interaction and morphology
• Some Recent Problems
• A number of filaments dont run perpendicular to
  the Galactic plane
• Zeeman and Faraday measurements plus synchrotron
  lifetime
• Anisotropy of scatter broadened OH/IR stars
• Diffuse non-thermal emission places a limit of
  pervasive field lt 10mmG
• Diffuse X-ray emission due to ICS with
  Tdust30K requires B10 mG
• Filaments not representative of the large-scale
  Field

21
A Turbulent Model of the Nonthermal Radio
Filaments

• The mean field is weak but strong turbulent
  activity amplifies the field locally
• Turbulent energy is two orders of magnitude
  higher than the disk
• High velocity dispersion
• Hyper-scattering medium
• Analogous to hydro turbulence with vorticity and
  MHD simulations
• B increases, produces filamentary structure
  until it reaches equipartition
• Generation
• Amplification rate eddy turnover time scale
  106-7 yrs
• The length of the filament outer scale of
  turbulence 10s pcs
• Expulsion
• Turbulent region is confined in GC, the field is
  expelled by turbulece
• Diffusion time scale t(diffusion) L2 / h 10
  times the eddy turnover time scale
• In the disk
• The region not confined, t(diffusion) gtgt eddy
  turnover time scale in the disk
• Turbulent energy is much smaller than in the GC

22

B field vectors of flux tubes (yellow and red)
Vortes tube (grey)
Nordlund et al. (1992)
23
Conclusions II

• ICS of the seed photons from dust cloud can
  explain some of the diffuse X-ray features in the
  GC
• Although non-thermal radio filaments may have
  strong magnetic field,
• they can not be representative of the
  large-scale filed in the GC
• A turbulent dynamo model in a highly turbulent
  region of the GC may explain the origin of
  nonthermal radio filaments

24
6 and 1.2cm VLA images of the Galactic Center
25

• The dispersion plot is minimum at 20min
• An expanding self-absorbed synchrotron source
  with a delay of 20min implies plasma ejection
  took place 54min before the 7mm peak (van der
  Laan 1966).
• No near-IR or sub-millimeter data)
• Continuous ejection?

26
SMT Observations (870micorn)March Campaign
27
Submillimeter 7mm and 0.45mm Emission (CSO)
March Campaign
28
IR (1.6-1.9mm) vs. Radio (7mm) (September
Campaign)
29
Radio (7mm) vs X-ray (September Campaign)
30
IR (1.6-1.9mm) vs. X-Ray (September Campaign)
31
Radio (7mm) vs X-ray (March Campaign)
40, 56 or 65min delay
32
Spectrum of Sgr A
Sgr A
33
(No Transcript)
34
Radio (7mm) vs X-ray (September Campaign)
35
Power Spectrum

• Assuming the P 1.3h variability is due to
  circular motion
• a/M0 no spin
• (r/M)orbit (P/2 p M)2/3 6Rsch
• (for comparison rISCO 2Rsch
• The same scale size as the region where sub-mm
  emission is expected to peak

36
(No Transcript)
37
Light Curve of SgrA at 1.6, 1.87 and 1.90 microns

• The variation is similar in all three filters
• Great instrument because of background and PSF
  stability