10 - IN2P3

Propagation of Galactic
Cosmic Rays
Antje Putze
LAPTh/LAPP
DMAstroLHC (ANR-12-BS05-0006)
1
What do we know about
nuclear cosmic rays today?
Nuclear cosmic rays are
•
energetic particles
[Bothe & Kolhörster, Z. Phys. (1929)]
•
of extraterrestrial origin
[Hess, Sitzungsberichte der kaiserl.
Akademie (1912)]
•
bombarding Earth’s
atmosphere, and
•
producing atmospheric
particle showers
[Auger et al., Rev. Mod. Phys. (1939)]
[Lee & Kirby, Marvel Comics (1961)]
2
Energy spectrum
extends over 12 orders of
magnitude in energy and 32 of
magnitude in intensity,
•
can be described by a simple
power law:
dN (E)
/E
dE
•
has three major features:
•
the knee at ~ 4.5 ∙ 1015 eV
•
the ankle at ~ 4 ∙ 1018 eV
•
a cut-off at ~ 4 ∙ 1019 eV
3
Galactic Cosmic Rays
•
Solar Modulation
The cosmic-ray spectrum
Composition & abundances
Spallation
[Hörandel, Advances in Space Research (2008)]
•
87% Hydrogen
•
12% Helium
•
1% heavier nuclei
similar abundances to those in the solar
system, but overabundances for elements
Z = 3 — 5, 20 — 25, …
4
Spatial distribution of
Galactic cosmic rays
NG
C4
63
1
610 MHz
Mi
lky
1412 MHz
408 MHz
[Ekers & Sancisi, A&A (1977)]
[Haslam et al., A&AS (1982)]
radio halo (few kpc) due to cosmic rays around the galactic disc
➞ galactic diffusive halo
5
Wa
y
secondaries
primaries
6
1. Sources & Acceleration
diffusive shock acceleration
secondaries
primaries
6
1. Sources & Acceleration
diffusive shock acceleration
II. Propagation in the ISM
secondaries
primaries
6
diffusion, convection,
re-acceleration
1. Sources & Acceleration
diffusive shock acceleration
II. Propagation in the ISM
secondaries
diffusion, convection,
re-acceleration
primaries
III. Solar System & Detection
solar modulation,
geomagnetic cut-off
6
q
@N
@t
=
~ · (K̂xx rN
~
q + r
|{z}
|
{z
~c N ) + @ p2 Kpp @ 12 N
V
@p
@p p
}
|
{z
}
sources
spatial di↵usion & convection
momentum di↵usion
h
⇣
⌘ i
@p
p ~ ~
@
1
1
N
r
V
N
N
c
@p @t
3
⌧s
⌧r N
|{z}
|{z}
|
{z
}
spallation
energy losses
radioactive decay
Cosmic-ray transport equation
7
Diffusion Model
Diffusion equation becomes solvable assuming a cylindrical geometry of the
Galaxy with 2 zones: the galactic disc & the diffusive halo
•
Semi-analytical approach e.g. USINE @ lpsc.in2p3.fr/usine
✅ fast computation
⛔️ simplified description of the
interstellar medium
•
q
Numerical approach
e.g. GALPROP @ galprop.stanford.edu
✅ data based description of the
interstellar medium
⛔️ very slow
8
Parameters and observables
The most important parameters
are linked to
•
•
•
p,
the acceleration mechanisms
injection spectrum:
Q(R) / qR ↵
the propagation mechanisms
K(R) / K0 R
diffusion:
VC
convection:
VA
re-acceleration:
He
,…
B/C
,…
10
Be 9
/B
e, …
the geometry of the Galaxy
diffusive halo size:
L
9
lpsc.in2p3.fr/cosmic-rays-db
Constraining propagation
models
sophisticated propagation models
precise experimental data
US
IN
AM
S
E
q
sophisticated statistical tools
MC
MC
parameters
PDF
10
observables
6 publications [AP, Coste, Derome, Donato, Maurin, Perotto, Taillet (2009 - 2014)]
Which model is the best?
IMP7-8
Voyager1&2
0.3
Be/9Be
Same results for
0.35
10
B/C
Diffusion models with re-acceleration and/or convection preferred, but
al.: An
MCMC technique to sample transport and sour
diffusion slope δ varies fromA. Putze
0.3etto
0.8
ACE 97-98
0.6
0.5
ACE 01-03
AMS-01
0.25
0.4
HEAO-3
ACE 97-9
ISOMAX
Spacelab-2
B/C
CREAM 04
0.3
0.15
✅ abundant
✅ elemental separation needed
0.2
Standard DM II
0.1
Standard DM III
[AP, Derome, Maurin, A&A (2010)]
Modified DM III
Secondary-to-primary ratio26
(TOA)
Al/27Al
0
10-1
!
IS
10
102
3
Standard DM II
ISEE-3
He/4He (TOA)
Standard DM III
0.25
0.12
0.2
0.1
0
10-1
3
10
Ek/n (GeV/n)
Balloon
0.14
• 3He/4He
1
0.1
Cl/Cl
0.05
0.25
36
•
0.2
IMP7-8
Balloon #
Balloon #
Balloon #
ISEE-3
Ulysses
Voyager1
Voyager1&2
Balloon72 (400 MV)
Balloon73
(500 MV)
Modified
DM III
Balloon77+Voyager (400 MV)
Balloon89 (1400 MV)
IMP7+Pioneer10 (540 MV)
ISEE3 (740 MV)
ISEE3(HIST) (500 MV)
Voyager87 (360 MV)
Ulysses
IMAX92 (750
MV)
SMILI-I (1200 MV)
ACE 97-98
SMILI-II (1200 MV)
AMS-01 (600 MV)
BESS98 (700 MV)
CAPRICE98 (700 MV)
0.2
0.15
Ulysses
0.08
✅ very abundant
⛔️ isotopic separation needed
0.15
ACE 97-
0.1
0.06
0.1
0.04
[Coste, Derome, Maurin, AP, A&A (2012)]
0 -2
0
10
10-1
11
0.05
Model II with prior (φ = 700 MV)
Model III with prior (φ = 700 MV)
Model II (φ = 700 MV)
Model III (φ = 700 MV)
0.05
0.02
IS
10-1
1
1
10
10
102
102
103
Ek/n [GeV/n]
Ek/n (GeV/n)
0
10-1
Fig. 10. Shown are the envelopes of 68% CL (shaded areas) and best-
How big is the diffusive halo?
L determines the amount of
dark matter in cosmic rays!
0.05
L [kpc]
10
Be/9Be
Al/27Al
36
Cl/Cl
10
Be/9Be + 26Al/27Al + 36Cl/Cl
26
0.04
first PDF of L from an MCMC
analysis with radioactive
secondaries
very sensitive to the LISM
0.03
Model
too
fewIII
rh =data
0
precise
0.02
0.01
[AP, Derome, Maurin, A&A (2010)]
0
40
60
80
100
120
140
160
5
!
φ = 520 MV
4.5
MED
4
3.5
L [kpc]
first direct exclusion of small
values of L with low-energy
secondary positrons
sensitive to solar modulation
δ
δ
δ
δ
+
3
sec. e
+
1.3 × sec. e
2.5
2
1.5
[Lavalle, Maurin, AP, arXiv:1407.2540]
1
ex
B/C included
δ = 0.5
δ = 0.6
δ = 0.7
δ = 0.85
clu
de
d
MIN
-6.5
-6
-5.5
log(K /L ⋅ Myr/kpc)
0
12
= 0.5
= 0.6
= 0.7
= 0.85
-5
-4.5
α
What the primaries tell us…
AMS 01
2.55
BESS 98
BESS-TeV
source slope 2.25 ⩽ ⍺ ⩽ 2.5 for
diverse propagation models
source slope ⍺ similar for all
primaries Z = 1, …, 26
2.50
protons
helium
68% CL
95% CL
2.45
2.40
2.35
2.30
2.25
!
A
➞ universality of the
acceleration mechanism
B
C
D
different propagation
models
13
[AP, Maurin, Donato, A&A (2011)]
What about systematics?
Precise cosmic-ray measurements give small statistical uncertainties,
and systematic uncertainties from model ingredients are dominating!
600
2
K0 [kpc /Myr]
400
200
0
0.005
0.01
Parameter estimation already very tricky in a simple configuration…
14
[Maurin, AP, Derome, A&A (2010)]
WIMPs
Indirect dark matter
searches
“exotic”
primaries
secondaries
primaries
charged cosmic-ray channels: e+, p̄, d,̄ …
15
Positrons — difficult probes
for dark matter searches
pu
lsa
Well modelled with
•
secondaries:
rs
✅ diffusion models
⛔️ uncertainties on propagation
parameters
•
primaries:
DM
✅ pulsars, dark matter
annihilation/decay, acceleration of
secondaries in sources, …
⛔️ very large uncertainties
⛔️ large boost factor needed for
dark matter interpretation
but no unique interpretation…
16
[Boudaud, AP, et al., A&A (in preparation)]
Antiprotons — strong
constraints for dark matter
Recent antiproton measurements in very good agreement with a
pure secondary origin
antiproton flux [GeV m2 s sr]-1
10-1
10-2
10-3
AMS (M. Aguilar et al.)
10-4
BESS-polar04 (K. Abe et al.)
BESS1999 (Y. Asaoka et al.)
BESS2000 (Y. Asaoka et al.)
10-5
CAPRICE1998 (M. Boezio et al.)
CAPRICE1994 (M. Boezio et al.)
PAMELA
10-6
10-1
1
10
102
kinetic energy [GeV]
G. 1: The antiproton energy spectrum at the top of the payload obtained in this work compared
17
h contemporary measurements [21–25] and theoretical calculations for a pure secondary pro-
[Adriani et al., PRL (2010)]
Anti-deuterons — the future?
excellent target for indirect dark matter searches due to its
high signal-to-noise ratio [Donato, Fornengo, Salati, PRD 2000]
•
•
high production threshold of
~17 mp
dark matter
production at rest
astrophysical
✅ dark matter annihilation
⛔️ astrophysical production
•
low binding energy
✅ disintegration
⛔️ energy losses
[Donato, Fornengo, Maurin, PRD 2008]
No cosmic-ray anti-deuteron detected so far, but GAPS is coming soon…
18
New data is coming …
from
AMS-02
Alpha Magnetic Spectrometer
experiment installed on the ISS
since 19 mai 2011
direct measurement elemental/
isotopic fluxes of matter and antimatter, 108 < E < 1012 eV
Principal sub-detectors and their specificities
TRD: electron-proton separation
TOF: trigger, velocity and charge
Tracker: rigidity, charge sign, charge
RICH: velocity and charge
ECAL: electron-proton separation, energy
19
Conclusion
•
Current propagation models
suffer from large uncertainties
on ingredients
What you put in is what you get
out…
•
More and more precise
cosmic-ray data will be
available soon
Are you hunting for dark matter?
Need for better models/ingredients
•
•
Cosmic rays are
complementary and
competitive with collider and
direct dark matter searches
20
Your dark matter candidate
should reproduce all the
available data
➞ global fits
➞ GAMBIT
ℒ
What is GAMBIT?
Global And Modular BSM Inference Tool
•
Flexible and modular design for easy addition of new
•
models
•
experimental data sets and limits
•
scanning algorithms
ℒ
•
Frequentist or Bayesian methods with customisable
•
likelihoods
•
priors
•
nuisance parameters
•
A smart and fast LHC likelihood calculator
•
Plug’n’Play swapping of physics tools, scanners, likelihoods…
•
Open source release
21
gambit.hepforge.org
ℒ
Who is GAMBIT?
ℒ
25 Members — 14 Institutes
P. Athron, C. Balázs, T. Bringmann, A. Buckley, J. Conrad, J. Cornell, L.-A. Dal, J. Edsjö, B. Farmer, L. Hsu, P. Jackson, A. Krislock, A. Kvellestad, F.N. Mahmoudi, G. Martinez, M. Pato, A. Putze, A. Raklev, C. Rogan, A. Saavedra, C. Savage, P. Scott (PI), N. Serra, C. Weniger, M. White
10 Experiments
AMS-02, ATLAS, CTA, DARWIN, CDMS, DM-Ice, Fermi-LAT,
H.E.S.S., IceCube, LHCb
22