gamma-ray spectra from dark matter [email protected] tev particle astrophysics, venice,...

Torsten Bringmann, SISSA/ISAS

Scu

ola

Inter

nazionale Superiore di Studi Avanzati

- ma per seguir virtute e conoscenza

-

Gamma-ray spectra from dark matterannihilations

Torsten Bringmann, SISSA/ISAS & INFN (Trieste)

[email protected]

TeV Particle Astrophysics, Venice, 27-31 August 2007

Gamma-ray spectra from dark matter annihilations – p.1/17


Scu

ola

Inter



-

Indirect DM detectionThe basic idea:

γ

ν

e+

DM

DM

γ

e

p_

+

Dark matter has to be (quasi-)stable against decay...

...but can usually pair-annihilate into SM particles.

These annihilation products can then potentially bespotted in cosmic rays of various kinds.

The challenge: a clear discrimination against backgroundand astrophysical sources.



Scu

ola

Inter



-

Why gamma rays ?

Rather high rates

Almost no attenuation when propagating through the halo

Point directly to the sources

No assumptions about diffusive halo necessary



Scu

ola

Inter



-

Why gamma rays ?

Rather high rates

Almost no attenuation when propagating through the halo

Point directly to the sources

No assumptions about diffusive halo necessary

Clear spectral signatures to look for



Scu

ola

Inter



-

γ rays from DM annihilations

The expected gamma-ray flux [GeV−1cm−1s−1sr−1] from asource with a high DM density ρ is given by

dΦγdEγ

(Eγ,∆ψ) =〈σv〉ann

8πm2χ

∑

f

BfdNf

γ

dEγ︸︷︷︸

particle physics

·

∫

∆ψ

dΩ

∆ψ

∫

l.o.sdℓ(ψ)ρ2(r)

︸︷︷︸

astrophysics

≃(D2∆ψ

)−1 ∫

d3rρ2(r)

〈σv〉ann : total annihilation cross section

mχ : DM particle mass (for WIMPs: 50GeV . mχ . 5TeV)

Bf : Branching ratio into channel f

Nfγ : Number of photons per annihilation

∆ψ : angular resolution of detector

D : Distance to point-like source



Scu

ola

Inter



-

DM annihilation spectra

3 types of contributions:

Secondary photons from fragmentation of decay productsmainly through π0

→ γγ

results in a rather featureless spectrum

Line signals from χχ→ γγ, Zγ,Hγ

necessarily loop-suppressed: O (α2)

“smoking gun” signature

Final state radiation (FSR)appears whenever charged final states are present, O (α)

characteristic signature, usually dominant at high energies



Scu

ola

Inter



-

Secondary photons

Quark and gauge boson fragmentation give essentiallydegenerate photon spectra: (Figs. from Bertone et al., astro-ph/0612387)

0.001

0.01

0.1

0.1

1

1

10

100

1000

0.02 0.05 0.2 0.5

x = E/mχ

dNγ/dx

rescale

−→dNf, res

γ

dx(x) ≡ Af

dNfγ

dx(Bfx)

N.B.: Bf ∼ 1 − 1.50.001

0.01

0.1

0.1

1

1

10

100

1000

0.02 0.05 0.2 0.5x = E/mχ

dN

res

γ/dx

∼ x−1.5



Scu

ola

Inter



-

Direct annihilation into photons

The direct annihilation into photons (χχ→ γγ, Zγ,Hγ) resultsin very sharp line signals (natural width ∼ 10−3 due to Dopplershift).

10-14

10-13

10-12

10-11

10-10

10-9

10-8

10-7

10-6

10-5

10-4

Ph

oto

n F

lux

[cm

-2s-1

GeV

-1]

12 3 4 5 6

102 3 4 5 6

1002 3 4 5 6

1000

Photon energy [GeV]

Neutralino continuum gamma ray flux towards galactic centre -

NFW model, ∆Ω=10-3

sr

Background, Eγ-2.7

300 GeV neutralino added 50 GeV neutralino added

Fig. from Bergström, Ullio & Buckley ’97

but:energy resolution (& 10%) andsensitivity of current detectors inmany cases not sufficient to dis-criminate the signal from the con-tinuum part.

(A particularly prominant exception is, e.g.,

the case of almost pure Winos and Higgsi-

nos, see Hisano et al. ’05.)



Scu

ola

Inter



-

Final state radiation(internal bremsstrahlung)

Whenever DM annihilates into charged final states f , thisprocess is automatically accompanied by χχ→ ffγ.

For mf ≪ mχ, the spectrum is usually dominated byphotons emitted collinearly from the charged final states

→ spectrum rather model-independent.

Under the following circumstances, however, photonsradiated from charged virtual particles can dominate:

t-channel annihilation into bosonic f

a symmetry violated by ff but not by ffγ

→ these contributions are highly model-dependent.



Scu

ola

Inter



-

Collinear photons

propagator for f :

∝ 1(k+p)2−m2

f

= 12k·p

k

p

For collinear photons, the virtual f is almost on-shell→ Logarithmic enhancement of the cross section (x ≡ Eγ/mχ):

dNdx ∼ σ(χχ→ ff) · αQ

2

π F(x) log sm2

f

(1 − x)

(see, e.g., Birkedal et al., hep-ph/0507194)



Scu

ola

Inter



-

Collinear photons

propagator for f :

∝ 1(k+p)2−m2

f

= 12k·p

k

p

For collinear photons, the virtual f is almost on-shell→ Logarithmic enhancement of the cross section (x ≡ Eγ/mχ):

dNdx ∼ σ(χχ→ ff) · αQ

2

π F(x) log sm2

f

(1 − x)

(see, e.g., Birkedal et al., hep-ph/0507194)

Example: LKP in UEDmB(1) ∼ 1 TeVhigh branching ratiointo leptons (∼ 60 %)

Bergström et al., PRL ’05a

x2dN

eff

γ/dx

x = Eγ/mB(1)

1

0.1

0.1

0.03

0.01

0.01

qq

qq, τ+τ−

total



Scu

ola

Inter



-

Charged virtual particles (1)

“Light” charged bosonic final states get an enhancement fromt-channel diagrams if the internal particles are degenerate inmass with the DM particles:

M ∝ 1k1·p1

1k2·p2

≈ 1m2

χE1E2

high Eγ small E1 or E2

(Note that the contraction of fermion final legs

leads to an additional Ef in the numerator)



Scu

ola

Inter



-


“Light” charged bosonic final states get an enhancement fromt-channel diagrams if the internal particles are degenerate inmass with the DM particles:

M ∝ 1k1·p1

1k2·p2

≈ 1m2

χE1E2

high Eγ small E1 or E2

(Note that the contraction of fermion final legs

leads to an additional Ef in the numerator)

Example: HiggsinoTeV masshigh W+W− b. r.

Bergström et al., PRL ’05b

Eγ [TeV]

d(σv) γ/dE

γ[1

0−

29cm

3s−

1TeV

−1]

0.1

1

10

102

103

104

0.1 0.5 1 2

mχ = 1.5 TeVW+W− : 0.39



Scu

ola

Inter



-


The 3-body final state may be al-lowed by a symmetry that is notsatisfied fot the 2-body final state.

Example: Leptons in SUSYusually helicity suppressedsuppression no longer efficientfor an additional photon in thefinal state, with Eγ ∼ mχ

Bergström, PLB ’89

even greater enhancementwhen sleptons degenerate withneutralino! → mSUGRA...



Scu

ola

Inter



-

FSR spectra

provide a unique and distinct signature (not possible tomistake for astrophysical processes)

the cutoff is more pronounced than for secondary photons DM mass can be determined to a higher accuracy

Can even be used to distinguish between different DMcandidates!

Example:B(1) vs. Higgsino(assume same mass and

energy resolution of 15 %)

Eγ [TeV]

E2 γ

d(σv) γ/dE

γ[1

0−

29cm

3s−

1TeV

]

10

102

103

0.1 0.5 1 2

Bergström et al., astro-ph/0609510

LSP

LKP



Scu

ola

Inter



-

FSR and SUSYTB, Bergström & Edsjö, ’07 (in prep.)

include FSR from all possible finalstates in DarkSUSY

Perform a scan over the whole MSSMinclude ∼ 106 models with Ωχh2 as determined by WMAP

log10mχ [GeV]

log10Zg/(

1−Zg)

FSR/sec.< 0.10.1 − 0.20.2 − 0.50.5 − 1.01.0 − 2.02.0 − 5.05.0 − 10.0> 10.0−7

−6

−5

−4

−3

−2

−1

0

1

2

2

3

3

4

4

5

log10mχ [GeV]

< 0.10.1 − 0.20.2 − 0.50.5 − 1.01.0 − 2.02.0 − 5.05.0 − 10.0> 10.0−7

−6

−5

−4

−3

−2

−1

0

1

2

2

3

3

4

4

5

FSR/(γγ + Zγ)

“Gaugino”

“mixed”

“Higgsino”

preliminary preliminary



Scu

ola

Inter



-

MSSM - total FSR fluxesTB, Bergström & Edsjö, ’07 (in prep.)

log10mχ [GeV]

log10Zg/(

1−Zg)

FSR

> 10−1210−13 − 10−1210−14 − 10−1310−15 − 10−1410−16 − 10−1510−17 − 10−1610−18 − 10−17< 10−18

7

−7

6

−6

5

−5

4

−4

3

−3

2

−2

1

−1

00

11

2

2

2

3

3

3

4

4

4

5

5

log10mχ [GeV]

> 10−1210−13 − 10−1210−14 − 10−1310−15 − 10−1410−16 − 10−1510−17 − 10−1610−18 − 10−17< 10−18

−7

−6

−5

−4

−3

−2

−1

0

1

2

2

3

3

4

4

5

γγ

preliminary preliminary

All fluxes given in photons/(cm2s)

FSR: flux above 0.6mχ



Scu

ola

Inter



-

MSSM - FSR componentsTB, Bergström & Edsjö, ’07 (in prep.)

2 3 4x

-7

-6

-5

-4

-3

-2

-1

0

1

2

3

4

5

y

l8

l7

l6

l5

l4

l3

l2

l1

Legend

2 3 4x

-7

-6

-5

-4

-3

-2

-1

0

1

2

3

4

5

y

l8

l7

l6

l5

l4

l3

l2

l1

Legend

2 3 4x

-7

-6

-5

-4

-3

-2

-1

0

1

2

3

4

5

y

l8

l7

l6

l5

l4

l3

l2

l1

Legend

2 3 4x

-7

-6

-5

-4

-3

-2

-1

0

1

2

3

4

5

y

l8

l7

l6

l5

l4

l3

l2

l1

Legend

2 3 4x

-7

-6

-5

-4

-3

-2

-1

0

1

2

3

4

5

y

l8

l7

l6

l5

l4

l3

l2

l1

Legend

2 3 4x

-7

-6

-5

-4

-3

-2

-1

0

1

2

3

4

5

yl8

l7

l6

l5

l4

l3

l2

l1

Legend

W+W− W+H− H+H−

ℓ+ℓ− uu tt



Scu

ola

Inter



-

FSR in mSUGRATB, Bergström & Edsjö, ’07 (in prep.)

log10mχ [GeV]

log10Zg/(

1−Zg)

W+W−/sec.< 0.10.1 − 0.20.2 − 0.50.5 − 1.01.0 − 2.02.0 − 5.05.0 − 10.0> 10.0−7

−6

−5

−4

−3

−2

−1

0

1

2

2

3

3

4

4

5

log10mχ [GeV]

< 0.10.1 − 0.20.2 − 0.50.5 − 1.01.0 − 2.02.0 − 5.05.0 − 10.0> 10.0−7

−6

−5

−4

−3

−2

−1

0

1

2

2

3

3

4

4

5

τ+τ−/sec.

Almost degenerate stops, typical in mSUGRA models, can give rise toenormous FSR contributions even for rather low neutralino masses!



Scu

ola

Inter



-

Summary

Final state radiation

in many situations completely dominates the spectrum forEγ & 0.6mχ (not only for heavy DM particles!)

provides unique and distinct spectral signatures

allows a precise determination of the DM mass due to thepronounced cutoff

can even be used to distinguish between different DMcandidates

should be regarded as at least equally

important for the indirect detection of DM

as line signals!


gamma-ray spectra from dark matter [email protected] tev particle astrophysics, venice,...

Documents