1
galactic cosmic rays andthe turbulent heliospheric tail
Paolo Desiati1,2 & Alexander Lazarian2
1 WIPAC - Wisconsin IceCube Astrophysics Center2 Department of Astronomy
University of Wisconsin - Madison
Midwest Magnetic Fields Workshop, Madison, WIApril 4th, 2012
Friday, April 6, 2012
Paolo Desiati
cosmic rays spectrum
≈E-2.7
≈E-3.1
≈E-2.7
• spectral structure & mass composition hold information on
‣ origin of cosmic rays
‣ propagation from sources to Earth
‣ anisotropy in arrival distribution
‣ energy dependence
‣ angular scale
2
1 GeV - 1 TeV 1 TeV - 1 PeV > 1 PeV extra-galactic
Friday, April 6, 2012
Paolo Desiati
cosmic ray acceleration in supernova remnants
SN1006
W. Baade & F. Zwicky, Physical Review 46, 76, 1934
3
• diffusive shock acceleration in galactic SNR (Baade & Zwicky, 1934 & Fermi, 1949)
φCR =cnCR(E)
4π
nCR(E) ≈ E−γ
RSN
2πR2d
· H
D(E) density of cosmic rays
cosmic ray flux
energy emitted by one SN
cosmic ray acceleration efficiency
radius of galactic disk
rate of supernovae in the Galaxy
propagation term
D(E) ∝ Eδ diffusion coefficient
φCR ≈ 2.4 ·�
ESN
1051erg
�
·�CR
·�
15kpc
Rd
�2
·�
RSN
30yr
�
·(γ − 2) · 3−δ
·�
E
1TeV
�−γ−δ
[TeV −1m−2s−1sr−1]
X-ray (Chandra)opticalradio
Friday, April 6, 2012
Paolo Desiati
cosmic rays observationsall-particle spectrum
4
p ≈ E-2.66
He ≈ E-2.58
PamelaAdriani et al. (2011)
CREAMAhn et al. (2010)
p ≈ E-2.80
He (×0.1) ≈ E-2.71
p ≈ E-2.67
He ≈ E-2.48
Friday, April 6, 2012
Paolo Desiati
cosmic rays observationsall-particle spectrum
5
residual all-particle spectrum
≈E-3.0
≈E-3.1
PamelaAdriani et al. (2011)
p ≈ E-2.80
He (×0.1) ≈ E-2.71
p ≈ E-2.67
He ≈ E-2.48
KASCADE-GrandeArtega-Velàzquez et al. (2010)
Gaisser & StanevPDG
Friday, April 6, 2012
Paolo Desiati
cosmic rays observationsanisotropy
6
05101520
α (hr)
24
0〫
60〫δ (deg)
05101520
α (hr)
24
0〫
50〫
δ (deg)
05101520
α (hr)
24
0〫
50〫
-50〫
δ (deg)
05101520
α (hr)
24
0〫
60〫
δ (deg)
equatorial coordinates
Nagashima et al. (1998)Hall et al. (1999)
ARGO-YBJZhang et al. (2009)
Tibet ASγAmenomori et al. (2006)
Super KamiokandeGuillian et al. (2007)
MilagroAbdo et al. (2009)
< 100 GeV 4 TeV
10 TeV
4 TeV
5 TeV
-90˚
α (hr)0510152024
IceCubeAbbasi et al. (2010)
20 TeV
Friday, April 6, 2012
Paolo Desiati
cosmic rays observationsanisotropy
7
equatorial coordinates relative intensity
Amenomori et al. (2011)
Abbasi et al. (2012)
Tibet-ASγ 5 TeV
IceCube-59 20 TeV
Friday, April 6, 2012
Paolo Desiati
anisotropy vs. energy
• CR anisotropy changes phase ~100 TeV
• global amplitude is modulated
Tibet ASγ Amenomori et al., ApJ. 626, L29, 2005
IceCube-22 Abbasi et al., ApJ, 718, L194, 2010
EAS-TOP Aglietta et al., ApJ, 692, L130, 2009
IceCube-59 Abbasi et al., ApJ, 746, 33, 2012
ARGO-YBJ Zhang 31st ICRC Łódź-Poland,2009
360˚
-90˚
0˚
-90˚
360˚ 0˚
20 TeV
400 TeV
IceCubeAbbasi et al. (2012)
δfluctuations =3
23/2
1π1/2
D(E)Hc
D(E) ∝ Eδ
8
Friday, April 6, 2012
Paolo Desiati
anisotropy vs. angular scale
detector acceptance
direct exp measurement+ detector acceptance
-0.2
-0.18
-0.16
-0.14
-0.12
-0.1
-0.08
-0.06
-0.04
-0.02
0
0.02
0.04
0.06
0.08
0.1
x 10-2
0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360
large vs small scale anisotropy
averaged modulation over a given angular rangelow angular gradient
fine structurehigh angular gradientacceptance-corrected
4 hr = 60˚
9
Friday, April 6, 2012
Paolo Desiati
2 hr = 30˚
4 hr = 60˚
IceCube
Milagro
20 TeV
1 TeV
10
Abbasi et al. (2011)
statistical significanceequatorial coordinates
cosmic rays observationsanisotropy
Abdo et al. (2008)
Friday, April 6, 2012
Paolo Desiati
origin of small scale anisotropy ?astrophysics
Salvati & Sacco, arXiv:0802.2181Drury & Aharonian, Astropart. Phys. 29, 420 (2008)• CR from Geminga: ~90-200 pc, 340,000 yr ago
• magnetic connection & propagation in turbulent LIMF
• anisotropic MHD turbulence in the ISM
‣ particles streaming along magnetic field lines over ~100 pc (from a source) interact with O(1pc) ISM turbulence
‣ pitch angle scattering peaked near the direction of LIMF
Salvati, Astron. & Astrophys. arXiv:1001.4947
Malkov et al., ApJ 721, 750, 2010
11
Friday, April 6, 2012
Paolo Desiati
origin of small scale anisotropy ?effect of turbulence
12
‣ diffusion regime breaks down within mean free path
‣ interaction with turbulent interstellar magnetic field
‣ assuming an underlying dipole anisotropy, fractional localized regions form the effect of magnetic field turbulence
‣ the residual maps provide an image of magnetic field turbulence < 10’s pc
‣ cosmic ray energy spectra might also be affected by this propagation effects
Giacinti & Sigl, arXiv:1111.2536
20° smoothing
10 PeV
10 PeV
50 PeV
Friday, April 6, 2012
Paolo Desiati
diffusive propagation models ...
• … assume uniform diffusion coefficient across the Galaxy
• … do not account for energy-dependent interaction with ISM turbulence
• … do not account for magnetic field geometry
• … cannot explain non-dipolar anisotropy structures
• … break down within mean free path
13
Friday, April 6, 2012
Paolo Desiati
from the Galaxy to our local interstellar medium
Milky Way
Benjamin (2008)
Frisch
Frisch
< 30,000 pc >
< 500 pc >
< 10-50 pc >
< 0.001 - 0.05 pc >
Local Bubble
Local Interstellar Cloud
Heliosphere
14
Friday, April 6, 2012
Paolo Desiati
100 AU100 AU > 1,000 - 10,000 AU
~30
0-60
0 A
U
~1.4 TeV ~1.4 TeV ~14 - 140 TeV
~ 5
- 1
0 Te
V
3 µG
15
Rg ∼10−3
Z
�E
1 TeV
� �µG
B
�pcthe heliosphere
and the LIMF
Pogorelov & Zank (2004)
Vinterstellar flow ~ 26 km/s ≳ VAlfén
Friday, April 6, 2012
Paolo Desiati16
the heliospheremagnetic structure3D simulations of heliosphere Opher et al., arXiv:1103.2236 !
3D simulation of heliosphere/heliotailPogorelov et al., ApJ, 696, 1478, 2009
~150 AU ~1,000’s AU
~200 AU~0.1-1 AU
Friday, April 6, 2012
Paolo Desiati
• the wake downstream the interstellar flow develops turbulence from plasma velocity difference across the heliopause (similar to Kelvin-Helmholtz instability)
• charge-exchange processes decelerate the solar wind near the heliopause, producing an effective drag force that pushes the higher ISM density into the heliosheath. This generates Rayleigh-Taylor instability oscillations with amplitude 10’s AU over 100’s years - Liewer et al. (1996).
• charge-exchange processes in plasma-neutral fluid model produces alternate growing and damping of Alfvénic, fast and slow turbulence modes, with amplitude 10-100 AU and slowly propagating downstream along the heliopause - Shaikh & Zank (2010).
‣ The 10-100 AU turbulent ripples propagate outward the ISM and are damped by ion-neutral collisions in mfp ~ 300 AU - Spangler et al. (2011).
the heliosphereturbulence
17
Friday, April 6, 2012
Paolo Desiati
scattering on heliosphericturbulence
18
Friday, April 6, 2012
Paolo Desiati
scattering on heliosphericturbulence
19
Friday, April 6, 2012
Paolo Desiati
• cosmic rays > 100 TeV do not feel the influence of the heliosphere
• cosmic rays < 100 TeV are influenced by the heliosphere from the downstream region
• resonant scattering of 1-10 TeV cosmic rays with 100’s AU turbulence ripples re-organizes the arrival direction distribution
• cosmic rays streaming along the LIMF experience the largest effect from the downstream region, and a minimal effect upstream
• perpendicular scattering is critical and determines the gradient region in cosmic ray arrival direction distribution
‣ evaluations and calculations to verify this scenario
scattering on heliosphericturbulence
20
Friday, April 6, 2012
Paolo Desiati21
PD & Lazarian, arXiv:1111.3075
magneticequator
Funsten et al. (2009)Schwadron et al. (2009)Heerikhuisen et al. (2010)
LIMF direction compatible with• Ca II absorption & H I lines, Frisch (1996)• radio emission from inner heliosheath, Lallement et al. (2005), Opher et al. (2007)• polarization measurements, Frisch (2010)
scattering on heliosphericturbulence
Friday, April 6, 2012
Paolo Desiati22
PD & Lazarian, arXiv:1111.3075
magneticequator
Funsten et al. (2009)Schwadron et al. (2009)Heerikhuisen et al. (2010)
LIMF direction compatible with• Ca II absorption & H I lines, Frisch (1996)• radio emission from inner heliosheath, Lallement et al. (2005), Opher et al. (2007)• polarization measurements, Frisch (2010)
scattering on heliosphericturbulence
!
Friday, April 6, 2012
Paolo Desiati
spectral feature associated to anisotropyAbdo A.A. et al., Phys. Rev. Lett., 101, 221101 (2008)
harder spectrum in region A
2 hr = 30˚ 2 hr = 30˚
Milagro & ARGO-YBJ
harder than average spectrum from region A
γ < 2.7 at 4.6 σ levelEc = 3 - 25 TeV
similar to hardening of “diffuse” cosmic rays by Pamela, CREAM, ATIC-2
23
Friday, April 6, 2012
Paolo Desiati
origin of spectral hardening ?
24
‣ magnetic polarity reversals due to the 11-year solar cycles compressed by the solar wind in the magneto-tail
‣ turbulence makes reconnection fast and not affected by ohmic dissipation
‣ magnetic mirror @ single reconnection as site of acceleration (test particle)
Lazarian & PD, ApJ, 722, 188, 2010
!Sweet (1959) Parker (1957)
Lazarian & Vishniac, ApJ, 517, 700, 1999
Friday, April 6, 2012
Paolo Desiati
origin of spectral hardening ?
25
‣ magnetic polarity reversals due to the 11-year solar cycles compressed by the solar wind in the magneto-tail
‣ turbulence makes reconnection fast and not affected by ohmic dissipation
‣ magnetic mirror @ single reconnection as site of acceleration (test particle)
Lazarian & PD, ApJ, 722, 188, 2010
~ 0.5 - 6 TeV
Friday, April 6, 2012
Paolo Desiati
Conclusions
• < 100 TeV cosmic ray anisotropy generated by interaction with the very local interstellar medium
• scattering with turbulence inside and in the outer heliospheric boundary to play an important role to explain large scale and small scale TeV cosmic ray anisotropy
• might explain change of cosmic ray anisotropy between 20 TeV and 400 TeV
• spectral hardening observed by Milagro & ARGO-YBJ from the downstream direction from re-acceleration of a fraction of cosmic rays in stochastic magnetic reconnection within the heliotail
• similar hardening observed by Pamela and CREAM could be related to the heliotail, although astrophysical explanations @ source and from propagation are possible
26
Friday, April 6, 2012