NOW 2014 Conca Specchiulla – 8.09.2014

Nicolao Fornengo Department of Physics (Theory Division) – University of Torino

and Istituto Nazionale di Fisica Nucleare (INFN) – Torino Italy

fornengo@to.infn.it nicolao.fornengo@unito.it

www.to.infn.it/~fornengo www.astroparticle.to.infn.it

Torino

Astrophysical Interpretation of AMS-02 Leptonic Data

Report on: Di Mauro, Donato, NF, Lineros, Vittino, JCAP 04 (2014) 006 [arXiv:1402.0321]

Cosmic-rays leptons

�  Excellent data on cosmic-rays leptons are available from space-borne detectors, from about up 0.5 GeV to few hundreds of GeV

-  Flux

-  Flux

-  Flux

-  PF

10 GeV the positron fraction decreases with increasingenergy as expected from the secondary production ofcosmic rays by collision with the interstellar medium.The positron fraction is steadily increasing from 10 to!250 GeV. This is not consistent with only the secondaryproduction of positrons [17]. The behavior above 250 GeVwill become more transparent with more statistics whichwill also allow improved treatment of the systematics.

Table I (see also [13]) also presents the contribution ofindividual sources to the systematic error for different binswhich are added in quadrature to arrive at the total system-atic uncertainty. As seen, the total systematic error at thehighest energies is dominated by the uncertainty in themagnitude of the charge confusion.

Most importantly, several independent analyses wereperformed on the same data sample by different studygroups. Results of these analyses are consistent with thosepresented in Fig. 5 and in Table I (see also [13]).

The observation of the positron fraction increase withenergy has been reported by earlier experiments: TS93[18], Wizard/CAPRICE [19], HEAT [20], AMS-01 [21],PAMELA [22], and Fermi-LAT [23]. The most recentresults are presented in Fig. 5 for comparison. The accu-racy of AMS-02 and high statistics available enable thereported AMS-02 positron fraction spectrum to be clearlydistinct from earlier work. The AMS-02 spectrum has theunique resolution, statistics, and energy range to provideaccurate information on new phenomena.The accuracy of the data (Table I and [13]) enables us to

investigate the properties of the positron fraction withdifferent models. We present here the results of comparingour data with a minimal model, as an example. In thismodel the eþ and e# fluxes,!eþ and!e# , respectively, areparametrized as the sum of individual diffuse power lawspectra and the contribution of a single common sourceof e$:

!eþ ¼ CeþE#!eþ þ CsE

#!se#E=Es ; (1)

!e# ¼ Ce#E#!e# þ CsE

#!se#E=Es (2)

(with E in GeV), where the coefficients Ceþ and Ce#

correspond to relative weights of diffuse spectra for posi-trons and electrons, respectively, and Cs to the weight ofthe source spectrum; !eþ , !e# , and !s are the correspond-ing spectral indices; and Es is a characteristic cutoff energyfor the source spectrum. With this parametrization thepositron fraction depends on five parameters. A fit to thedata in the energy range 1–350 GeV based on the numberof events in each bin yields a "2=d:f: ¼ 28:5=57 and thefollowing: !e# # !eþ ¼ #0:63$ 0:03, i.e., the diffusepositron spectrum is softer, that is, less energetic withincreasing energy, than the diffuse electron spectrum;!e# # !s ¼ 0:66$ 0:05, i.e., the source spectrum isharder than the diffuse electron spectrum; Ceþ=Ce# ¼0:091$ 0:001, i.e., the weight of the diffuse positron fluxamounts to !10% of that of the diffuse electron flux;Cs=Ce# ¼ 0:0078$ 0:0012, i.e., the weight of the com-mon source constitutes only !1% of that of the diffuseelectron flux; and 1=Es ¼ 0:0013$ 0:0007 GeV#1, corre-sponding to a cutoff energy of 760þ1000

#280 GeV. The fit isshown in Fig. 6 as a solid curve. The agreement betweenthe data and the model shows that the positron fractionspectrum is consistent with e$ fluxes each of which is thesum of its diffuse spectrum and a single common powerlaw source. No fine structures are observed in the data. Theexcellent agreement of this model with the data indicatesthat the model is insensitive to solar modulation effects[24] during this period. Indeed, fitting over the energyranges from 0.8–350 GeV to 6.0–350 GeV does not changethe results nor the fit quality. Furthermore, fitting the datawith the same model extended to include different solarmodulation effects on positrons and electrons yields simi-lar results. This study also shows that the slope of the

num

ber o

f tria

ls

number of positrons

posi

tron

frac

tion

(b)(a)

0

5

10

15

450 500 550 6000.11

0.112

0.114

0.116

0.118

0.12

positron fraction0.10 0.11 0.12 0.13

num

ber o

f tria

ls

0

50

100

150

FIG. 4 (color). (a) Stability of the measurement in the energyrange 83.2–100 GeVover wide variations of the cuts fitted with aGaussian of width 1.1%. (b) The positron fraction shows nocorrelation with the number of selected positrons.

1 10 210

AMS-02

-110

PAMELAFermi

FIG. 5 (color). The positron fraction compared with the mostrecent measurements from PAMELA [22] and Fermi-LAT [23].The comparatively small error bars for AMS are the quadraticsum of the statistical and systematic uncertainties (see Table Iand [13]), and the horizontal positions are the centers ofeach bin.

PRL 110, 141102 (2013) P HY S I CA L R EV I EW LE T T E R Sweek ending5 APRIL 2013

141102-7

e�

e+

e+/(e� + e+)

(e� + e+)

PAMELA Collab, Nature 458 (2009) 607 [arXiv:0810.4995] PAMELA Collab, PRL 111 (2013) 081102 [arXiv:1308.0133] PAMELA Collab, PRL 106 (2011) 201101 [arXiv:1103.2880] Fermi LAT Collab, PRL 108 (2012) 011103 [arXiv:1109.0521] Fermi LAT Collab, PRD 82 (2010) 09004 [arXiv:1008.3999] AMS-02 Collab, PRL 110 (2013) 141102 AMS-02 Collab, 33rd ICRC Conference (2013) AMS-02 Collab, 33rd ICRC Conference (2013)

From: AMS-02 Collab, PRL 110 (2013) 141102

Sources of cosmic-rays leptons

�  Primaries

- Supernova remnants (SNR): electrons - Pulsar Wind Nebulae (PWN): electrons, positrons

- Exotic sources, like e.g. Dark Matter (electrons, positrons)

�  Secondaries

- From cosmic-rays (mostly p and He) interactions in the Galaxy: (electrons, positrons)

In this talk: astrophysical interpretation

�  Primaries

- Supernova remnants (SNR): electrons - Pulsar Wind Nebulae (PWN): electrons, positrons

- Exotic sources, like e.g. Dark Matter (electrons, positrons)

�  Secondaries

- From cosmic-rays (mostly p and He) interactions in the Galaxy: (electrons, positrons)

Primary electrons from SNR

�  Particles accelerated through first-type Fermi mechanism

�  Electron source spectrum:

�  Normalization estimated from radio data:

�  Green catalog: 274 SNR with information on distance, age, radio flux and radio spectral index

Q(E) = Q0

✓E

E0

◆��

exp

✓� E

Ec

◆

Q0 = 1.2 · 1047 GeV�1 (0.79)�

d

kpc

�2 h ⌫

GHz

i(��1)/2

B

100µG

��(�+1)/2 B⌫r

Jy

�

Ec = 2 TeV

Primary electrons from SNR

�  We subdivide SNR into:

- Near component: -  closer than 3 kpc -  35 SNR in catalog, treated as single, independent sources -  and derived from radio data

- Far component: -  treated as an single average population with:

-  and as free parameters in the fit

⇢(r, z) = ⇢0raexp (�r/r0) exp (�|z|/z0)

(a , r0 , z0) = (2.35 , 1.5 kpc , 0.1 kpc)

Lorimer, IAU Symposium 218 (2004) 105 [astro-ph/0308501]

�

Q0�

Q0

Primary electrons and positrons from PWN

�  Pulsars can be treated as sources of highly-boosted pair-produced electrons and positrons

�  Electron/positrons source spectrum:

�  Normalization derived from the spin-down energy W0

�  ATNF catalog: PWN with information on distance, age and

total emitted power

�  Free parameters:

Ec = 2 TeV

Z 1

Emin

dE E Q(E) = ⌘W0 W0 ⇡ ⌧0E

✓1 +

tage⌧0

◆2

Q(E) = Q0

✓E

E0

◆��PWN

exp

✓� E

Ec

◆

PSR decay time ⌧0 ⇠ 10 kyr

�PWN ⌘

Secondary electrons and positrons

(p , He)prim + ISM �! e±

101

102

103

104

100 101 102 103 104

(T)2

Φ [

Ge

V/n

ucl

eo

n m

-2 s

-1 s

r-1]

T [GeV/nucleon]

p and He

p AMS-02He AMS-02

p fitHe fit

IS p fitIS He fit

Dedicated fit on AMS-02 data (*)

(*) Haino (ASM-02 Collab), 33rd ICRC Conference (2013) (*) Choutko (ASM-02 Collab), 33rd ICRC Conference (2013)

Transport equation

diffusion energy losses source term

Geometry: cylindrical diffusive halo (R, L) + thin disk (h = 100 pc)

Diffusion: uniform in the whole (disk + diffusive halo) volume

Energy losses: synchrotron on magnetic fields, inverse Compton on ISRF Transport in the Heliosphere: Force field

K(E) = K0 � (R/1 GV)�

@tN � ~r ·hK(E)~rN

i+ @E

⇥b

loss(E)N⇤= Q(E, ~x, t)

Model L (kpc) � K0 (kpc

2Myr

�1)

MED 4 0.70 0.0112

�

10-3

10-2

10-1

100 101 102 103 104

E3 \ [G

eV2 c

m-2

s-2

sr-1

]

E [GeV]

e-

3m bandII

PWNSNR local

SNR d > 3 kpcTOT IS

TOTAMS-02

PAMELAFERMI

AMS-01AMS02 Prel.

10-4

10-3

10-2

100 101 102 103 104

E3 \ [G

eV2 c

m-2

s-2

sr-1

]

E [GeV]

e+

3m bandII

PWNTOT IS

TOTAMS-02

PAMELAFERMIHEAT

CAPRICEAMS02 Prel.

10-3

10-2

10-1

100 101 102 103 104

E3 \ [G

eV2 c

m-2

s-2

sr-1

]

E [GeV]

e+ + e-

3m bandII

PWNSNR local

SNR d > 3 kpcTOT IS

TOTAMS-02FERMIHESS

AMS02 Prel.

10-2

10-1

100

100 101 102 103

e+ /(e+ +e

- )

E [GeV]

e+/(e++e-)

3m bandII

PWNTOT IS

TOTAMS-02

PAMELAFERMI

AMS-01HEAT

CAPRICEAMS02 Prel.

Fit to AMS-02 data

electrons positrons

sum positron fraction

Fit to AMS-02 data

�  Reconstructed parameters (for MED)

- Far SNR:

- PWN:

- Normalization of secondaries:

⌘ = 0.0320± 0.0016�PWN = 1.90± 0.03

Q0 = (2.748± 0.027)⇥ 1050 GeV�1

� = 2.382± 0.004

qsec = 1.080± 0.026

� = (830± 22) MV

The case for pulsars energy cut-off

10-4

10-3

10-2

100 101 102 103 104

E3 Φ

[G

eV

2 c

m-2

s-2

sr-1

]

E [GeV]

e+

II3σ band Ec=1 TeV

3σ band Ec=10 TeVPWN Ec=1 TeV

PWN Ec=10 TeVTOT Ec=1 TeV

TOT Ec=10 TeVAMS-02

PAMELAFERMI

10-2

10-1

100

100 101 102 103

E3 Φ

[G

eV

2 c

m-2

s-2

sr-1

]

E [GeV]

e+/(e++e-)

II3σ band Ec=1 TeV

3σ band Ec=10 TeVPWN Ec=1 TeV

PWN Ec=10 TeVTOT Ec=1 TeV

TOT Ec=10 TeVAMS-02

PAMELAFERMI

positrons positron fraction

Ec = 10 TeVEc = 10 TeV

Ec = 1 TeVEc = 1 TeV

The high-energy window and local sources

�  The high-energy sector is dominated by local discrete sources:

-  The electron and (electron+positron) fluxes are well compatible with a local SNR remnant contribution

-  The positron flux (and the positron fraction) require local positron sources, and PWN from the ATNF catalog can provide a fairly viable solution

�  We therefore attempt two additional analyses to study the high-energy positron window (relevant also for DM):

- “Single-source” analysis - “Powerful-sources” analysis

“Single-source” analysis

�  We explore the PWN parameter space by injecting positrons from a single PWN-like emitter

�  Free parameters: -  spectral index -  distance -  age -  emitted power in the e+/e- channel

�  Once we find the required properties (if any) of a PWN-like emitter to reproduce the AMS-02 data, we look back into the ATNF catalog

�PWN

dT⌘W0

“Single-source” analysis

101

102

103

104

105

10-1 100

T [

kyr]

d [kpc]

γ = 1.5

ηW0 = [1046,1047] ergηW0 = [1047,1048] ergηW0 = [1048,1049] ergηW0 = [1049,1050] ergηW0 = [1050,1051] ergPWN ATNF cataloguePWN near and young

“Single-source” analysis

101

102

103

104

105

10-1 100

T [ky

r]

d [kpc]

γ = 1.8

ηW0 = [1046,1047] ergηW0 = [1047,1048] ergηW0 = [1048,1049] ergηW0 = [1049,1050] ergηW0 = [1050,1051] ergPWN ATNF cataloguePWN near and young

101

102

103

104

105

10-1 100

T [ky

r]

d [kpc]

γ = 2.0

ηW0 = [1046,1047] ergηW0 = [1047,1048] ergηW0 = [1048,1049] ergηW0 = [1049,1050] ergPWN ATNF cataloguePWN near and young

101

102

103

104

105

10-1 100

T [ky

r]

d [kpc]

γ = 2.2

ηW0 = [1046,1047] ergηW0 = [1047,1048] ergηW0 = [1048,1049] ergPWN ATNF cataloguePWN near and young

“Single-source” analysis: best-fit PWN

Name �fit dfit Tfit ⌘W0,fit �2/dof dcat Tcat W0,cat ⌘fit(kpc) (kyr) (10

49erg) (kpc) (kyr) (10

49erg)

Geminga 1.74 0.24 344.6 0.341 0.68 0.25 342 1.25 0.27

J1741-2054 1.68 0.25 378.0 0.413 0.62 0.25 386 0.47 0.88

B1742-30 1.52 0.19 539.1 0.770 0.54 0.2 546 0.83 0.92

J1918+1541 1.65 0.64 2355 6.48 0.92 0.68 2310 3.4 1.90

“Single-source” analysis: fluxes

10-3

10-2

10-1

100 101 102 103 104

E3 Φ

[G

eV

2 c

m-2

s-2

sr-1

]

E [GeV]

e-

IIGeminga

SNR localSNR d > 3 kpc

TOTAMS-02

PAMELAFERMI

AMS-01HEAT

CAPRICE

10-2

10-1

100

100 101 102 103

e+/(

e++

e- )

E [GeV]

e+/(e++e-)

IIGeminga

TOTAMS-02

PAMELAFERMI

AMS-01HEAT

CAPRICE

10-5

10-4

10-3

10-2

10-1

100 101 102 103 104

E3 Φ

[G

eV

2 c

m-2

s-2

sr-1

]

E [GeV]

e+

IIGeminga

TOTAMS-02

PAMELAFERMIHEAT

CAPRICE

10-3

10-2

10-1

100 101 102 103 104

E3 Φ

[G

eV

2 c

m-2

s-2

sr-1

]

E [GeV]

e++e-

IIGeminga

SNR localSNR d > 3 kpc

TOTAMS-02FERMI

ATICHEAT

CAPRICEBETSHESS

electrons positrons

sum positron fraction

“Powerful-source” analysis

�  We identify in the ATNF catalog a limited number of PWN potentially able to sizeably contribute to the local high-energy positron flux

-  For definiteness: 5 “most-powerful”sources -  Selected by a ranking algorithm, according to their contribution to

the local positron flux

�  For each of them, we allow and to vary, and we fit the AMS-02 data

�PWN ⌘

“Powerful-sources” analysis (5)

10-2

10-1

100

10-1

Fra

ctio

n

η

GemingaJ2043+2740

B0355+54Monogem

J0538+2817

10-2

10-1

100

1.4 1.5 1.6 1.7 1.8 1.9 2 2.1 2.2

Fra

ctio

n

γPWN

GemingaJ2043+2740

B0355+54Monogem

J0538+2817

efficiency spectral index

ATNF Association d[kpc] T [kyr] W0[1049

erg]

J0633+1746 Geminga 0.25 343 1.26

J2043+2740 1.13 1204 26.0

B0355+54 1 567 4.73

B0656+14 Monogem 0.28 112 0.178

J0538+2817 1.3 622 6.18

frac

tion

“Powerful-source” analysis (5 + 1)

10-2

10-1

100

10-1

Fra

ctio

n

η

GemingaJ2043+2740

B0355+54Monogem

J0538+2817PWN background

10-2

10-1

100

1.4 1.5 1.6 1.7 1.8 1.9 2 2.1 2.2

Fra

ctio

n

γPWN

GemingaJ2043+2740

B0355+54Monogem

J0538+2817PWN background

efficiency spectral index

frac

tion

5 “most-powerful” pulsars + “PWN background”

“Powerful-source” analysis (5+1)

10-4

10-3

10-2

10-1

100

100 101 102 103 104

E3 Φ

[G

eV

2 c

m-2

s-2

sr-1

]

E [GeV]

e-

IISNR local

SNR d > 3 kpcPWN TotGeminga

J2043+2740B0355+54Monogem

J0538+2817TOT

AMS-02PAMELA

FERMI

10-2

10-1

100

100 101 102 103

e+/(

e++

e- )

E [GeV]

e+/(e++e-)

IIPWN TotGeminga

J2043+2740B0355+54Monogem

J0538+2817TOT

AMS-02PAMELA

FERMI

10-5

10-4

10-3

10-2

10-1

100 101 102 103 104

E3 Φ

[G

eV

2 c

m-2

s-2

sr-1

]

E [GeV]

e+

IIPWN TotGeminga

J2043+2740B0355+54Monogem

J0538+2817TOT

AMS-02PAMELA

FERMI

10-3

10-2

10-1

100 101 102 103 104

E3 Φ

[G

eV

2 c

m-2

s-2

sr-1

]

E [GeV]

e++e-

IISNR local

SNR d > 3 kpcPWN TotGeminga

J2043+2740B0355+54Monogem

J0538+2817TOT

AMS-02FERMI

electrons positrons

sum positron fraction

Summary and Conclusions

�  The whole AMS-02 leptonic data have been analysed in a self-consistent theoretical framework -  Primary electrons: distant sources ave flux + local SNR in Green catalog -  Secondary leptons: from cosmic-rays interactions with the ISM -  Primary high-E positrons: PWN from ATNF catalog

�  Remarkable agreement with the data for all types of analyses -  average contribution from all ATNF catalog PWN -  single dominant PWN in ATNF -  a few most-powerful PWN in ATNF

�  The analysis provides hints on the required properties of emitters, which will require further cross-checks in other channels (like radio or gamma-rays) and theoretical models

Backup slides

103 10410-25

10-24

10-23

10-22

10-21

MDM @GeVDsv@cm

3 êsecD

DM DM Æ mm, Iso profile

PAMELA+ AMS e+

FERMI +HESS e±

FERMI 5oâ 30o

FERMI 3oâ 3o

FERMI Dwarfs

103 10410-25

10-24

10-23

10-22

10-21

MDM @GeVD

sv@cm

3 êsecD

DM DM Æ mm, NFW profile

PAMELA+ AMS e+

FERMI +HESS e±

FERMI 5oâ 30o

FERMI 3oâ 3o

FERMI Dwarfs

Positrons: DM interpretation

Cirelli, Kadastik, Raidal, Strumia, NPB813 (2009) 1 as updated in arXiv:0809.2409v5

bounds from gamma-rays on “positive” interpretation of positron fraction