Radio galaxy physics at low frequencies: lobes, jets and environments

1
Radio galaxy physics at low frequencieslobes,
jets and environments

• Martin Hardcastle(University of Hertfordshire,
  UK)
• LFRU, 9th December 2008
• With thanks to collaborators including Judith
  Croston, Joanna Goodger, Dharam Vir Lal, Leslie
  Looney, Nazirah Jetha, Irini Sakelliou

2
Outline

• Why work at low frequency? Constraints from and
  on the physics of radio galaxies
• Low-frequency radio and environments (a
  radio-galaxy perspective)
• Future prospects LOFAR, LWA and SKA

3
Radio galaxy physics

• Jets of relativistic electrons/positrons
  magnetic field emitted at relativistic speeds
  from central black hole.
• Interaction of jets with environment gives rise
  to kpcMpc scale structures.
• (This talk only covers kpc-scale structures.)
• Goal is to figure out physics of/physical
  conditions in these.

4
Hotspot
Jet
Core
Lobe
Hotspot
Plume
FRI
FRII
5
Radio galaxy physics

• Two key emission processes are
• Synchrotron radiation (relativistic electrons
  magnetic fields) peak frequency ?c goes as B?2,
  total emissivity as B2 ?2. For B 1-10 nT,
  ?103-104 electrons give rise to GHz-freq radio
  emission
• Synchrotron appears in all wavebands from radio
  through to X-ray. (From above, higher frequencies
  gt higher energies of electrons gt shorter loss
  timescales.)

6
Radio galaxy physics

• Two key emission processes are
• Inverse-Compton scattering (relativistic
  electrons and background photon field, e.g. the
  CMB or the optical AGN emission). Peak frequency
  goes as ?photon?2, total emissivity as Uphoton
  ?2. For CMB, ?103 electrons scatter to 1-keV
  X-ray photons
• Inverse-Compton is seen in optical, X-ray and
  above (no significant low-frequency photon
  background to scatter).

7
Radio galaxy physics

• Synchrotron emission from a single electron is a
  smooth, peaked function of frequency.
• So a power-law distribution of electrons gives a
  power-law synchrotron spectrum.
• If spectrum is power-law, why go to the trouble
  of working at low frequencies?

J(?) goes as F(x) where x ?/?c
8
Energetics

• Energy density in lobes dominated by electrons
  with lowest energies since p 2.
• In principle observing at low frequencies takes
  us towards lower energies so probes
  numerically/energetically dominant population.
• In practice lowest-energy ee- radiate at 100s of
  kHz in lobes/plumes and will never be observable
  (barring exotic techniques like measuring the S-Z
  decrement).
• Not a convincing argument for low frequency in
  itself, so what is?

9
Inverse-Compton

• Detection of inverse-Compton emission in
  principle allows us to measure the magnetic field
  strength and therefore electron number density,
  total energy etc.
• But CMB inverse-Compton X-rays at 1 keV come from
  electrons with ? 1000, radiating at 10s of MHz
  in a B-field of 0.3 nT. (Different for SSC.)

Inverse-Compton emission from the lobes of Pictor
A (MJH Croston 2005)
10
Inverse-Compton

• To infer magnetic field strengths with confidence
  from radio X-ray data we need radio
  observations that either constrain the relevant
  electron population or at least allow a good
  extrapolation to be made.
• (X-ray imaging currently only out to 10 keV gt
  100s of MHz but see later.)
• Low-freq observations crucial (e.g. Konar poster)

Inverse-Compton emission from the lobes of 3C353
(Goodger 2008)
11
Inverse-Compton

• B-fields measured and close to equipartition in
  the lobes and hotspots of FRIIs (MJH 02
  Croston 05).
• Variation in B-field strength measured in a few
  cases.
• No IC detection yet for FRI sources.

B/Beq for lobes in 3C FRIIs, detections and
non-detections (Croston 05)
12
Low-energy cutoff

• Power-law approximation breaks down if there is a
  low-energy cutoff in the electron energy spectrum
  (i.e. ?min gtgt 1).
• Below frequencies corresponding to this energy we
  will see a transition to the ?1/3 power law of
  the tail of the single-electron spectrum (details
  will depend on whether there is a true cutoff
  unlikely in practice).

13
Low-energy cutoff

• Best hope of seeing this in hotspots of FRII
  sources where B-field is highest.
• Possibly seen in a few sources Cyg A (Carilli
  91, Lazio 06), 3C123 (MJH 01) may be
  constrained by optical inverse-Compton, if
  detectable (e.g. Brunetti 2002, MJH 2002).
• High-resolution low-frequency data needed for
  more objects.

14
Spectral ageing / injection index

• Since the energy loss rate of electrons goes as
  ?2, energy loss timescale (E/(dE/dt)) goes as
  1/? higher-energy electrons lose energy faster
  than lower-energy ones.
• On some realistic assumptions about pitch angle
  diffusion, we obtain a predicted synchrotron
  spectrum which is steeper at higher frequencies.

15
Spectral ageing

• In this picture low-frequency observations are
  crucial because they provide the view of the
  source least affected by ageing.
• Many difficulties inherent in method
• Injection index is poorly known, may not be
  constant in FRIIs, no theoretical expectation
  that it should be in FRIs (though see Young
  2005).
• Adiabatic expansion effectively mimics spectral
  ageing, is a required process
• From IC work we know there are strong
  point-to-point B-field variations that are not
  taken into account in spectral ageing models.
• Any in situ acceleration invalidates assumptions.
• See later talks/posters for more discussion.

16
Spectral ageing

• Sometimes we have to use spectral ageing because
  there is no alternative.
• At least qualitatively there is some evidence
  that it works quantitatively spectral ages are
  of the order of dynamical ages where these can be
  determined.
• Large range of frequencies is important
  remember ? goes as ?1/2 so even 100 MHz 100 GHz
  is only 1.5 orders of magnitude in ?. Low-freq
  observations needed.

3C388 at 610 MHz, 1.4 GHz, 8.4 GHz and 90 GHz.
Note how jet, core and hotspot structures come to
dominate at higher frequencies.
17
Doppler boosting and source selection

• Easy to overlook the crucial role of
  low-frequency radio in source selection
• Since cores, jets and hotspots are strongly
  affected by beaming and flat-spectrum,
  high-frequency selection is biased.
• Low-frequency selection lets us select samples on
  quasi-isotropic properties.

5-GHz images of quasars from Bridle 1994
18
Environments

• Many other talks will discuss use of
  low-frequency observations in groups/clusters.
• Here we will concentrate on uses of low-frequency
  data in constraining interactions with
  environments of pre-existing radio sources.

19
Seeing the source properly!

• Low-frequency work has often been required to
  give us a complete view of the radio source.
• E.g. M87, (Owen 00), Hydra A (Lane 04)
• Prerequisite for environment studies

20
A caveat

• We need to be careful to distinguish between
  situations where low frequency is key and where
  sensitive, short-baseline observations at any
  frequency are all thats required.
• Example

21
3C278 at 1-arcsec resolution at 5 GHz with the
VLA (A, B, C configurations). (Machacek 07)
22
Caveat
3C278 at 6-arcsec resolution with the GMRT at
610 MHz (left) and the VLA at 5 GHz with
D-configuration data included (right). The
extended structure here did not require low
frequency for its detection (though it is
steep-spectrum) but simply short baselines!
23
Lower frequencies gt larger scales
3C40 in Abell 194 at L-band (left) and P-band
(right). X-ray contours overlaid. Low-frequency
data are needed to see the full extent of many
tailed radio sources therefore to assess their
environmental impact (Sakelliou 08)
24
Particle content problem

• No IC emission in FRIs gt must use external
  environments to constrain internal physical
  conditions. (Groups/clusters well imaged in
  X-ray at low redshift.)
• While FRIIs seem to be close to pressure balance
  with their environments with only ee- and
  B-fields, equipartition field strengths often
  imply pint ltlt pext in FRIs gt equipartition with
  ee- alone is wrong gt role for protons?
• Pressure estimates from radio data must be based
  on low-frequency observations to avoid serious
  bias.

25
Particle content problem
Some progress in understanding pressure balance
problem Croston 08
26
Particle content problem
Bridges
Plumes
Plumed sources seem systematically further out of
pressure balance at equipartition than bridged
ones favours a model where the pressure balance
is made up by entrained, heated thermal material
(Croston 08).
27
Life cycles and fuelling

• Low-frequency observations are sensitive to
  large-scale, old material gt allow us to trace
  previous AGN episodes.
• Interesting in context of
• Double-double FRIIs (later talks)
• Rare but interesting FRIs with evidence for
  multiple outbursts.

B2 083832A, Jetha 2008
28
Life cycles and fuelling

• With low-frequency observations we can select
  objects that are restarting from samples.
• Allows investigation of AGN duty cycle and
  fuelling mechanism for nearby, low-power sources
  (MJH 07 Jetha 08)

B2 083832A, Jetha 2008
29
The future

• GMRT and VLA are workhorse instruments now, but
  coming up we have
• LOFAR
• LWA
• SKA
• Plus combination at shorter wavelengths with
  e-MERLIN, EVLA, ALMA

LOFAR can already see 3C sources
30
Technical challenges

• At low frequencies we are going to see a large
  fraction of the sky can we make effective use of
  this to study bright sources?
• What about the time domain for radio-loud AGN?
• We will have to get used to a model in which we
  are not pointing a telescope and seeing only what
  we want to see e.g. for LOFAR radio-galaxy data
  may come primarily as a by-product of surveys (
  transients?) key science projects.

Early LOFAR observations
31
Lower frequencies

• Inverse-Compton work in particular will benefit
  from sensitive observations at 10s of MHz if
  its technically possible (particularly at high
  resolution) since few keV will remain the
  standard X-ray observing energy.
• At the same time, important synergy with
  higher-energy X-ray instruments allowing imaging
  of IC/CMB from electrons corresponding to 100-MHz
  electrons (e.g. Simbol-X).

32
Higher sensitivity/fidelity/dynamic range

• Current low-frequency instrumentation remains
  limited in image fidelity and dynamic range
  more so than at GHz frequencies.
• Next-generationtelescopes mustsolve this
  problemto allow wide-fieldhigh-sensitivityimagi
  ng.
• Work on brightsources withcomplex
  structurewill benefit.

3C353 VLAPT P-band Goodger 08
33
Multifrequency/polarization capability

• Faraday rotation in the diffuse hot medium will
  be a key complement to X-ray data possible to
  probe to much lower particle densities (cf.
  Tigran Arshakians talk).
• Sensitive, high-resolution broad-band
  polarization measurements at multiple, low
  frequencies required difficult but will
  revolutionize environment studies once we have
  them.

34
Large samples

• Existing imaging studies of large RG samples
  limited by observing time, available effort
• Largest samples with good radio data are 100
  objects
• Sensitivity will make it much easier to
  accumulate large samples
• Statistical studies of e.g. jet speeds, particle
  acceleration properties

35
Summary

• Although in recent years GHz-frequency imaging
  has dominated, low-frequency radio astronomy has
  much to offer to the study of the physics of
  radio-loud AGN.
• At present it offers the only way to study the
  extended regions of sources where environmental
  interactions are most important.
• New capabilities coming on-line soon will lead to
  exciting results.