Interstellar Scattering Joseph Lazio (Naval Research Laboratory) J. Cordes, A. Fey, S. Spangler, B. Dennison, B. Rickett, M. Goss, E. Waltman, M. Claussen,

Interstellar ScatteringInterstellar Scattering

Joseph Lazio(Naval Research Laboratory)

J. Cordes, A. Fey, S. Spangler, B. Dennison, B. Rickett, M. Goss, E. Waltman, M. Claussen, D. Jauncey, L. Kedziora-Chudczer, R. Ojha

Radio-wave Radio-wave ScatteringScattering

re ne ds

Electron density fluctuations

Refractive index fluctuations

Corrugated phase fronts Image distortions (cf.

atmospheric seeing) Characterized by a

scattering measureSM ne

2 dx

Ionized Interstellar MediumIonized Interstellar Medium

H II regions EM > 104 pc cm-6

Powered by O or B star(s)

Warm ionized medium (WIM)

n ~ 0.1 cm-3

T ~ 8000 K f ~ 0.2 1/6 of O star

luminosityWHAM survey

Scattering ObservablesScattering Observables Angular broadening

– Pulsars

– Extragalactic sources

– Masers and other Galactic sources

Intensity scintillations– Pulsars

– Extragalactic sources

Pulse broadening/scintillation bandwidth

Pulsars

(Spectral broadening)

Scattering characterized typically by scattering measure

SM ne2 dx

Not really scattering observables, but related observables include Rotation measure Optical emission from diffuse gas ( EM = ne

2 dx) Dispersion measure variations (DM = ne dx) Diffuse gamma-ray emission?

Radio-wave Scattering AnalysesRadio-wave Scattering Analyses

ne ds

~ 2 SM

Scattering physics– Density spectrum

• Spectral index

• Inner scale

– Scattering genesis

Distribution– “regional”

– Galactic

slope

coefficient SM

The Density Spectrum and Angular The Density Spectrum and Angular BroadeningBroadening

Point source at infinity

V(b) = e-D(b)/2

Phase structure function

D(b) = [(x) - (x+b)]2D(b) dq Pn(q)

[1–J0(bq)]

Pn(q) q- (or …)

Density Fluctuation Power Density Fluctuation Power SpectrumSpectrum

Density spectrum in local interstellar medium

Power law, with spectral index near Kolmogorov value– Notable exceptions!– Large dynamic range!

Interstellar plasma has large Reynolds number.

Turbulent processes responsible for density fluctuations(?).

Density spectrum elsewhere in Galaxy similar, probably.

Armstrong, Rickett, & Spangler (1995)

1 pc 1 AU

Extreme Scattering Extreme Scattering EventsEvents

Events simultaneous at 2.2 and 8.1 GHz

Duration of few weeks to months

intrinsic: Tb 1015 K

extrinsic: AU-sized refracting clouds in our Galaxy

ESE of 1741-038:ESE of 1741-038:1992 June 20 (18 cm)1992 June 20 (18 cm)

Need a new monitoring program!

Density Fluctuation Power Density Fluctuation Power SpectrumSpectrum

Armstrong, Rickett, & Spangler (1995)

1 pc 1 AUDensity spectrum in local

interstellar mediumPower law, with spectral

index near Kolmogorov value– Notable exceptions!– Large dynamic range!

Interstellar plasma has large Reynolds number.

Turbulent processes responsible for density fluctuations(?).

Density spectrum elsewhere in Galaxy similar, probably.

Turbulence Inner ScaleTurbulence Inner Scale If density fluctuations result

from turbulence, inner scale would be a dissipation scale.

Scattering resolved if b ~ /d.

Inner scale important if l1 ~ b. Inner scale estimates are

roughly 200 km. Spangler & Gwinn attribute it

to the ion inertial length or ion Larmor radius.

Note gap in coverage from 30 km to 1000 km.

Spangler & Gwinn (1990)

Sub-parsec magnetic fieldsSub-parsec magnetic fields

NGC 6334B and Cyg X-3 show rotation of image shape with frequency:– Different frequencies

sample different length scales in scattering medium.

Density fluctuations changing shape on these scales.

Magnetic fields aligning density fluctuations on this scale.

Yet Sgr A* and B1849+005 do not…

Scattering GenesisScattering Genesis

Scattering traces star formation– NGC 6334B (Trotter et al.)– Cygnus region (many

studies) Direct link more difficult

to establish– Spangler et al. vs. Simonetti

& Cordes and Spangler & Cordes

Should be able to do much better today and in future

2013+370/G74.9+1.2CTA 1

Where is the Scattering Medium?Where is the Scattering Medium?(“Regional”)(“Regional”)

Sources embedded in the medium are less scattered than background sources

Scattering must overcome the wavefront curvature.

Distance ambiguity for Galactic sources

No ambiguity for extragalactic sources

xgal = (DGC/GC) GC

Can solve for GC.

Where is the Scattering Medium?Where is the Scattering Medium?

B1849+005

B1849+005

PSR B1849+00

GC Scattered ImagesGC Scattered Images

Sgr A* displays enhanced angular broadening

OH/IR stars have maser spots with comparable diameters

GC scattering diameter: 1" @ 1 GHz

GC Scattering—Where?GC Scattering—Where?

Likelihood Results: xgal sources: GC < 500 pc OH masers:

50 pc < GC < 300 pc GC 150 pc xgal 75” @ 1 GHz Angular extent 1 (Note 1°

150 pc.) Inhomogeneous on  10–20 pc X-ray emitting gas + molecular

gas

Radial Extent of the WIMRadial Extent of the WIM(“Galactic”)(“Galactic”)

H I disks of nearby galaxies appear truncated

Due to extragalactic ionizing flux?

H II disk extends much farther?

Corbelli et al. 1989

H I

H (= H I + H II)

Radial Extent and Warp of the WIMRadial Extent and Warp of the WIM

WIM radial extent equals or exceeds H I:

H I disks truncated at R ~ 25–50 kpc (Galaxy a prototypical z = 0 Ly α cloud?)

C IV absorption toward H1821+643, R ~ 25 kpc

HVC models often require pressure support at R ~ 25 kpc

VLBA SurveyVLBA Survey

12 sources– 7 with |b| < 1°

– 5 with l ~ 180° and |b| < 10°

Cf. Dennison et al. 1984

Best-fit Radial ModelBest-fit Radial Model

No Perseus spiral arm

Perseus spiral arm at 25% of TC93

truncated disk

sech2 disk

Sources of ScatteringSources of Scattering

Truncated disk because of star formation?

Molecular clouds show radial truncation;

Star formation follows molecular clouds;

Scattering truncates where star formation does.

Similar to what is seen in other galaxies.

Molecular cloud distribution from CO survey by Wouterloot & Brand

26 kpc

Ne2001Ne2001(Cordes & Lazio 2002, astro-ph/0207156)(Cordes & Lazio 2002, astro-ph/0207156)

Number of data have nearly doubled.

Modifications from TC93:– GC component added; Diffuse component

truncated at 20 kpc;– Diffuse component made

thicker; Spiral arms extrapolated; Spiral arms made thicker;– Orion-Cygnus arm added;– Local Bubble and similar

regions added; “Clumps” and “voids”

added.

Anomalous Scattering EffectsAnomalous Scattering Effects

Multiple media can lead to anomalous scattering effects– Phase – Scattering angle 2

Effects occur because size of scattering region can become important in determining size of scattering disk.E.g., scattering of sources seen

through other galaxies. Important for LOFAR?

Infinitely extended scattering screen …

Or not.

Cosmic Rays, Cosmic Rays, rays, and the WIMrays, and the WIM

CRs are charged particles

Smooth CR energy spectrum

Magnetic irregularities scatter CRs

Same magnetic irregularities cause scattering?

1 pc 1 AU

CR energy spectrum/gyroradii

Summary: Interstellar ScatteringSummary: Interstellar Scattering

Exquisite probe of sub-parsec plasma physics– Density spectrum– Magnetic fields– Interstellar “clouds” (ESEs)– Cosmic rays?

Galactic distribution of scattering– Large-scale tracer of Warm

Ionized Medium (WIM)– Traces star formation

See also– Intraday variability– Pulsar parallax and proper

motions

VLBA itself has been an immense step forward.VLBA + other telescopes is good.

NMA will close gap around 100 km.

LOFAR will be wonderful instrument for scattering studies (2).Difficult to avoid scattering at

LOFAR frequencies! Space VLBI would be good,

if frequency is low enough. SKA will be even better.

FINISFINIS

GC Scattering—PulsarsGC Scattering—Pulsars

107–108 neutron stars: Massive star

formation High-energy sourcesSelection Effects:– beaming & LF– velocities– background– pulse broadening

GC Scattering—Pulse GC Scattering—Pulse BroadeningBroadening

GC ~ 350 seconds GHz-4

Periodicity search: long-period, shallow spectra pulsars,  > 8 GHz

Imaging search: steep-spectrum point sources, ~ 1” @ 1 GHz

10 seconds

Characterizing ScatteringCharacterizing Scattering

Strong scattering at SIRA (and LOFAR) frequencies:

• Fresnel radius RF = 3 x 1012 cm (D/100 pc)1/2(/1 MHz)-1/2

• rms phase in Fresnel radius >> 1

• Two characteristic regimes within strong scattering:o Diffractiveo Refractive

Rickett 1990

Refractive EffectsRefractive Effects

Unimportant time scales too long

• Refractive scintillation time scale -2

666

6660

tr (@ 1 MHz in yr)

Rickett et al. 1984

Diffractive EffectsDiffractive Effects

Diffractive scintillation seen commonly in pulsar observations at meter and centimeter wavelengths.

• Characteristic bandwidth, d ~ 3 kHz (@ 1 MHz)

• Characteristic time, td ~ 60 s (@ 1 MHz for v ~ 100 km/s)

No objects will scintillate (twinkle). Frequency

Tim

e

Scintille

Diffractive EffectsDiffractive Effects

• Pulse broadening smears out pulsar pulses.

• At SIRA frequencies, extreme pulse broadening can be obtained.

Most pulsars will not be seen as pulsed objects.

Diffractive EffectsDiffractive Effects• Angular broadening

distorts view of sources.

• Magnitude is large. Current SIRA specs

more than sufficient!

Local Bubble

Optical DepthOptical DepthElectrons responsible

for scattering also contribute to free-free optical depth.

0.24 MHz 0.4 MHz

Cosmic Rays and Cosmic Rays and raysrays

Diffuse -ray emission:

– p + p – e + p – e +

Cosmic Rays and Cosmic Rays and raysrays