T h e F e r mi G e V e x c e s s S i g n a l o - - PowerPoint PPT Presentation

1

T h e F e r mi G e V e x c e s s S i g n a l

• r

b a c k g r

• u

n d f

• r

D M s e a r c h e s ?

→ F . C a l

• r

e , I . C h

• l

i s & C W [ 1 4 9 . 4 2 ] , a c c e p t e d b y J C A P F . C a l

• r

e , I . C h

• l

i s , C . Mc C a b e & C W [ 1 4 1 1 . 4 6 4 7 ] , a c c e p t e d b y P R D S . C a r

• n

, A . A c h t e r b e r g , L . H e n d r i k s , R . R u i z d e A u s t r i & C W [ 1 5 2 . 5 7 3 ]

C h r i s t

• p

h We n i g e r

MIAPP Workshop, Munich, 24/02/2015

2

O v e r v i e w

 Introduction  History of the “Fermi GeV Excess”  Robust identification at higher latitudes  Implications & ways forward  Conclusions

3

I n t r

• d

u c t i

• n

4

I n d i r e c t S e a r c h e s f

• r

D a r k Ma t t e r ?

B-field Neutrinos

• Simple propagation
• But: hard to measure

Gamma rays

• Very simple propagation (geodesics)
• Absorption negligible on Galactic

scales

• Point towards their sources

Charged cosmic rays

• Electrons/positrons, nuclei
• Propagation distorted by

galactic magnetic fields

• Sizable energy losses &

interactions

Today's dark matter annihilation cross-section is roughly given by

Conditions during freeze-out are very different from today: The velocity averaged annihilation cross-sections can differ by orders of magnitude.

5

D a r k Ma t t e r S i g n a l F l u x

Characteristic Energy Spectrum Characteristic Morphology (point-like, extended or diffuse)

Particle Physics Astrophysics

Velocity averaged annihilation cross-section Photon energy spectrum per annihilation Dark matter mass Dark matter mass density It is convenient to define a “J-value”: Signal intensity:

[photon flux per steradian per energy]

Line-of-sight integral

[review DM searches with gamma rays: Bringmann & Weniger (2012)]

6

D M a n n i h i l a t i

• n

p r

• c

e s s e s

DM DM DM DM DM DM Gamma-ray lines: Two-body annihilation into photons Bremsstrahlung: Photon production in “hard process” Continuum emission: Photons from neutral pion decay DM DM Box-shaped spectra: Photons from cascade decay

7

Internal Bremsstrahlung (IB)

• radiative correction to

processes with charged final states

• Generically suppressed by

O(α) Gamma-ray lines

• from two-body

annihilation into photons

• forbidden at tree-leve,

generically suppressed by O(α²) Box-shaped spectra

• Cascade-decay into

monochromatic photons

• already at tree level

Continuum emission

E n e r g y s p e c t r u m

• f

p h

• t
• n

s f r

• m

D M a n n i h i l a t i

• n

8

The DM distribution very close (<1kpc) to the Galactic center is observationally

• nly poorly constrained.

Viable DM density profiles: Signal morphology (including substructure enhancement):

[Cirelli et al. (2010)]

A n a l y t i c a l D a r k ma t t e r d e n s i t y p r

• f

i l e s

[Pieri+ 0908.0195]

9

P

• t

e n t i a l t a r g e t s f

• r

s e a r c h e s wi t h p h

• t
• n

s

Galactic center (~8.5 kpc)

• brightest DM source in sky
• but: bright backgrounds

Dwarf Spheroidal Galaxies

• harbour small number of stars
• otherwise dark (no gamma-ray

emission) Galactic DM halo

• good S/N
• difficult backgrounds
• angular information

DM clumps

• w/o baryons
• bright enough?
• boost overall signal

Extragalactic

• nearly isotropic
• only visible close to

Galactic poles

• angular information
• Galaxy clusters?

Extended or diffuse: (for observations with gamma rays) Point-like: (for observations with gamma rays)

[review on N-body simulations: Kuhlen, Vogelsberger & Angulo (2012)]

Signal is approx. proportional to column square density of DM:

10

Galactic center

R e g i

• n

s

• f

i n t e r e s t ( ma x i mu m S / N )

Signal Background Region with optimal signal/noise: DM profile

11

G a l a c t i c c e n t e r a n a l y s i s

Tavakoli+ 2014; Gomez-Vargas+ 2014; Ackermann+ 2011; Hooper & Linden 2011 Constraints on the Galactic Halo Dark Matter from Fermi-LAT diffuse measurements [Ackermann+ 1205.6474] Constraints on WIMP Annihilation for Contracted Dark Matter in the Inner Galaxy with the Fermi-LAT [Gomez-Vargas+ 1308.3515]

12

S t a c k i n g d wa r f s

13

[Slide from Matthew Wood, 14.10.14 SLAC, presented on 5th Fermi Symposium] 100 GeV!!!

14

H . E . S . S .

• b

s e r v a t i

• n

s

• f

G a l a c t i c c e n t e r

Abramowski et al. 2011

For Atmospheric Cherenkov Telescopes, the backgrounds are completely dominated by unrejected electron and proton Crs → Isotropic backgrounds! Limits on a WIMP annihilation signal Flux from search region (green) compared to flux from background region (red). → the fluxes are consistent → upper limits on DM signal. `

Fermi dwarfs

Thermal cross-section Background Signal region

arXiv:1103.3266

Limits hold only for cuspy profiles!

15

T h e F e r mi G e V E x c e s s

16

F i r s t a p p e a r a n c e i n 2 9

First clear statements about properties of excess emission (morphology, spectrum etc, subject to some changes in later analyses): First very cautious comments by the LAT team, without any detailed characterization of the residual:

17

F

• l

l

• w

u p s t u d i e s

Goodenough & Hooper 2009 Hooper & Goodenough 2011 Hooper & Linden 2011 Boyarsky+ 2011 Abazajian & Kaplinghat 2012 Gordon & Macias 2013 Macias & Gordon 2014 Abazajian+ 2014 Daylan+2014 Hooper & Slatyer 2013 Huang+ 2013 Zhou+ 2014 Daylan+ 2014

At the Galactic center (roughly 7deg x 7deg) In the inner Galaxy (roughly |b|>1 deg to tens of deg)

[Daylan+ 2014] [Hooper & Slatyer 2013]

18

N

• n
• e

x p l a n a t i

• n

s

Neutral pion decay from proton interaction with molecular gas

Linden+ 1203.3539; Abazajian+ 1207.6047; Abazajian+ 1402.4090

• Requires proton spectrum with break to produce

spectrum

• Gives rise to wrong morphology

GeV emission from HESS GC HESS

Hooper & Goodenough 1010.2752; Boyarsky+ 1012.5839

• Emission is not a point source

Point source subtraction, general background systematics, etc

Boyarsky+ 1012.5939

• Proven not relevant in subsequent studies

Bremsstrahlung of CR electrons from GC on molecular gas

Yusuf-Zadeh+ 1206.6882; Gordon & Macias 1306.5725; Macias & Gordon 1312.6671; Abazajian+ 1402.4090

• Emission extends outside of the inner 250 pc (central

molecular zone), and is not correlated with molecular gas maps

Yusef-Zadeh+ 2012

21cm radio continuum, GBT

19

P

• s

s i b l e A s t r

• p

h y s i c a l i n t e r p r e t a t i

• n

s

Milli-second pulsars:

• Spectrum of known MSPs agrees reasonably

well with claimed GCE spectrum (except at sub-GeV energies)

• Observed luminosity function is claimed to be

incompatible with GCE (we don't see resolved MSPs at GC) Hooper+, Calore+, Cholis+ 2013

• Compatible with distribution of low-mass X-ray

binaries (possible MSP progenitors)

Cholis+ 2014

Abazajian, Canac, Horiuchi, Kaplinghat 1402.4090 Contribution from MSPs to GC flux, Wang, Jiang, Cheng astro-ph/0501245 Consistency of inner Galaxy observations with MSPs, Abazajian, 1011.4275 “Pulsars cannot account for the Inner Galaxy's GeV Excess”, Hooper, Cholis, Linden, Siegal-Gaskins, Slatyer, 1305.0830 Could be new population of MSPs, Mirabal, 1309.3428 Very broad population study, Yuan and Zhang, 1404.2318 Disk population cannot account for GC MSPs, Calore, Di Mauro, Donato, 1406.2706 Challenges for MSPs, Cholis, Hooper, Linden, 1407.5625 Population study and importance of ICS, Petrovic, Serpico, Zaharijas, 1411.2980 VHE gamma rays, Yuan and Ioka, 1411.4363 Statistical analysis, Mirabal, 1411.7410 LMXB population in M31, Voss, Gilfanov, astro-ph/0610649 Abazajian, Kaplinghat, 1207.6047 Gordon, Macias, 1306.5725

20

P

• s

s i b l e A s t r

• p

h y s i c a l i n t e r p r e t a t i

• n

s

Leptonic activity:

Petrovic+ 2014

• Recent injection of hard electrons at Galactic center,

~1 Myr ago

• Diffusion
• approx. spherical profile & emission

→

• Can “naturally” explain peaked spectrum
• The morphology, especially emission above 10 deg

(1.5 kpc) is hard to reproduce, since the energy loss time of electrons is < 1 Myr. Hadronic activity:

Carlson+ 2014

• Recent injection of protons at Galactic center,

100 Kyr – 2 Myr ago

• Diffusion causes approx. spherical CR

distribution, but target material is not spherical

21

D a r k Ma t t e r A n n i h i l a t i

• n

References: Goodenough & Hooper 0910.2998; Hooper & Goodenough 1010.2725; Hooper & Linden 1110.0006; Abazajian & Kaplinghat 1207.6047; Hooper & Slatyer 1302.6589; Gordon & Macias 1306.5725; Huang+ 1307.6862; Abazajian+ 1402.4090; Daylan+ 1402.6703; Abazajian+ 1410.6168; Calore+ 1409.0042; Calore+ 1411.4647; S. Murgia Fermi Symposium 2014

+ >O(100) models that explain the excess. Previous results were inconsistent

22

Q u e s t i

• n

s

The interpretation of the excess emission depends critically on:

• Excess energy spectrum
• Extension
• Spherically or other symmetry
• Uniformity of energy spectrum

→ How to estimate the associated uncertainties realistically?

23

R

• b

u s t c h a r a c t e r i z a t i

• n
• f

t h e G e V e x c e s s e mi s s i

• n

a t h i g h e r l a t i t u d e s

24

Do residuals look like this? Subtract 1) Known point sources 2) Diffuse foregrounds Fermi LAT; > 1 GeV

T h e G a l a c t i c h a l

• F
• r

e g r

• u

n d s u b t r a c t i

• n

Fermi bubbles

25

F e r mi P

• i

n t S

• u

r c e s & U n c e r t a i n t i e s

We mask all 2FGL point sources according to their relative contribution to the overall flux. [Ackermann+ 1501.02003] PSCs in the Galactic disk

26

D i f f u s e G a l a c t i c b a c k g r

• u

n d s

Inverse Compton scattering: Proton-proton collisions & subsequent neutral pion decay: low energy photon

• cosmic microwave background
• starlight

high energy electron

The diffuse gamma-ray emission from our Galaxy is produced by interaction of high energetic charged particles (electrons, protons, …) with the interstellar medium (mostly Hydrogen and Helium) and interstellar radiation field (Cosmic Microwave background, starlight, dust radiation)

high energy proton proton at rest

(at ~ sub-GeV energies also Bremsstrahlung)

27

T r a c e r s

• f

t h e i n t e r s t e l l a r me d i u m

[slides borrowed from I. Grenier 2010] Atomic neutral hydrogen Molecular hydrogen Dust reddening

28

S i mp l e s t d i f f u s e mo d e l c

• mp

a r e d t

• d

a t a

[slides borrowed from J.-M. Casandjian 2014]

29

T h e F e r mi B u b b l e s

Outflow at ~900 km/s estimated from quasar absorption lines (1kpc/Myr) Fermi Bubbles [Su+ 2010; Dobler+ 2010; Ackermann+ 2014] Ackermann+ 1407.7905 Fox+ 1412.1480 Possible explanations

• Jets from the black hole

[Guo & Mathews 2012, Yang+ 2012]

• Feedback from nuclear star formation

[Crocker & Aharonian 2011, Carretti+ 2013; Lacki 2014]

• Shocks from accretion flows onto Sgr A*

[Cheng+ 2011, Mou+ 2014]

• Spherical outflow from Sgr A*

[Zubovas+ 2011]

Have to be modeled by template.

30

T h e P r

• b

l e m' s G e

• me

t r y : C a r t

• n
• f

t h e Mi l k y Wa y i n d i f f u s e g a mma r a y s

Interstellar radiation field + magnetic field HI + HII + H2 Fermi Bubbles CR sources “Fermi GeV excess” ICS π0 (HI+H2) π0 (WIM) Bremss ~12% ~3% ~11% Emissivity along the line-of-sight: Fermi Bubbles CR electrons CR protons CR electrons:

• Inverse Compton emission (on ISRF)
• Bremsstrahlung (on gas)

CR protons:

• neutral pion decay (on gas)

F

31

Wh y l

• k

i n g a t h i g h e r l a t i t u d e s

GDE is mostly local!

32

S t r a t e g y

A.

We use Galactic diffuse emission models

• from literature
• extreme models

and perform standard template analysis

B.

We estimate uncertainties of these models by looking for residuals along the Galactic plane.

C.

We characterize all aspects of the excess emission from Galactic center, taking into account the empirically determined uncertainties.

[Ackermann+ 2012; 1202.4039] [Galprop v54]

(this was done before, though with a much more limited range of diffuse emission models)

33

S t u d y i n g s y s t e ma t i c u n c e r t a i n t i e s

• “Inner Galaxy”:
• We mask all point sources from the 2FGL

ROI: Components in the analysis: π0+Bremss free ICS free Bubbles constrained Isotropic constrained Excess template free Energy independent templates Energy dependent templates 2FGL fixed Fits independently in energy bins Spectral information from Galprop models → is neglected Galprop Galprop

34

R e l e v a n t p a r a me t e r s

Cosmic ray propagation

• Diffusion zone height: 4-10 kpc
• Diffusion constant: 2-60 1028 cm2/s
• Reacceleration: 0-100 km/s
• Convective winds: 0-500 km/s/kpc

Cosmic ray sources

• Distributed like SNR, pulsars, OB stars
• Electron/proton injection spectra

Cosmic-ray interaction

• ISRF: 50% variations
• Magnetic field: 5.8-117 μG @ GC
• Spin temperature: 150 K, optical thin
• Dust correction parameters

Diffusion equation:

Calculations done with: Galprop v54 http://galprop.stanford.edu/webrun

35

D e l i b e r a t e l i mi t a t i

• n

s

• f
• u

r a n a l y s i s

→ 60 representative GDE models

• no constraints from local CRs (although some models fit local CRs well)

36

T y p i c a l r e s i d u a l s f

• r
• n

e G D E mo d e l

• Left: Point source mask clearly visible
• Middle: Residuals at the level of <20% are observed
• Right: Readding the DM template clearly shows an extended

excess around the GC

37

T y p i c a l r e s i d u a l s

GCE template Longitude profile: Model A, 2.1-3.3 GeV

38

T y p i c a l r e s i d u a l s

Longitude profile: Model A, 2.1-3.3 GeV

39

C

• mp
• n

e n t s p e c t r a

Solid lines: model prediction (model A) GeV excess template

40

T h e

• r

e t i c a l mo d e l u n c e r t a i n t i e s

Spectra obtained after extreme variations of foreground model parameters In all cases, the excess template spectrum

• rises from 300 MeV to ~1 GeV
• peaks at 1-3 GeV
• falls power-law like above 3 GeV

(no cutoff at >10 GeV energies as previously claimed) Individual components in fit

• nly vary by ~O(2).

“Theoretical model uncertainties”

41

E mp i r i c a l mo d e l s y s t e ma t i c s : A n e s t i ma t e f r

• m

r e s i d u a l s i n t h e G a l a c t i c d i s k

ROI Relevant latitude range We can use Galactic disk as test region to estimate the impact of uncertainties in gas maps, modeled CR distribution, point source fits and masking, and instrumental effects

• n excess template fit at Galactic center.

Longitudinal variations photon sources are relatively mild. 2FGL We move the ROI and excess template along disk, and redo our fits.

42

F l u x i n e x c e s s t e mp l a t e s h i f t e d a l

• n

g t h e G a l a c t i c p l a n e

Control regions 1-11 (east disk) Control regions 12-22 (west disk) ROI

43

C

• v

a r i a n c e ma t r i x

• f

r e s i d u a l s p e c t r a

Residuals seen in the 24 energy bins and 22 test regions define a 24x24 covariance matrix: i, j = 1, …, 24; averaged over 22 test regions

44

P r i n c i p a l c

• mp
• n

e n t a n a l y s i s

Decomposition of covariance matrix: Principal components Random variables: Spectrum from different random components: Solid lines: measured Dashed lines: model

45

P r i n c i p a l c

• mp
• n

e n t a n a l y s i s

Solid lines: measured Dashed lines: model Normalization error <3% (from fit) Spectral slope error <0.01 (from fit) This can be understood in terms of small variations in the ICS and pi0 backgrounds. Variations in true ICS, pi0 flux: Corresponding over/undersubtraction is partially absorbed by GCE template

46

T h e

• r

e t i c a l v s . e mp i r i c a l mo d e l s y s t e ma t i c s

Empirical model uncertainties (yellow) and theoretical model uncertainties (blue lines) are significantly larger than the statistical error over the entire energy range. Have to take into account systematics to get meaningful results in spectral fits.

47

F i t s wi t h D M a n d a s t r

• s

p e c t r a

48

S a me p r

• c

e d u r e , b u t f

• r

t e n G C E s e g me n t s

49

I s t h e s p e c t r u m e v e r y wh e r e t h e s a me ?

A fit of DM bb spectra in each of the ten segments Results are consistent with hypothesis of one single spectrum at 95% CL!

• North/south symmetric
• East/west symmetric

50

H

• w

f a r d

• e

s t h e e x c e s s e x t e n d f r

• m

t h e G C ?

To explore the extension of the excess to high latitudes, we consider a hypothetical source with volume emissivity profile We find a lower limit on the extension of at least 1.48 kpc (corresponding to more than 10 degrees).

51

T h e h i g h e r

• l

a t i t u d e t a i l

• f

t h e F e r mi G e V e x c e s s

Gamma-ray intensity at 2 GeV:

• Most previous results agree within a factor of ~2, but disagree within error bars.
• The profile is compatible with the expectations from a DM annihilation signal

with contracted DM profile / power-law. No indications for radial cutoff.

Calore+ 2014

52

I mp l i c a t i

• n

s & Wa y s f

• r

wa r d

53

H i g h e r D M ma s s e s a r e a l l

• we

d

[see also Agrawal+ 2014] Annihilations into

• b-quarks give best fit, and is allowed up to DM masses around 73 GeV
• Higgs pair gives a good fit for DM mass around 125.7 GeV
• muons produce ICS emission that can be compatible with gamma-ray

data at |b|>2 deg, depending on the CR diffusion model

54

F i t s wi t h d a r k ma t t e r a n n i h i l a t i

• n

s p e c t r a

55

D wa r f s p h e r

• i

d a l l i mi t s

For cross-sections at the 95% CL exclusion limit from dwarf spheroidal galaxies, current dynamical and microlensing constraints still allow DM halo profiles that give rise to a signal morphology consistent with the observations.

56

D a r k Ma t t e r A n n i h i l a t i

• n

i n MS S M

LHC signatures:

• Chargino + Neutralino production
• Monojets
• Squark and gluino searches

Most relevant constraints from:

• Relic density
• LUX
• IceCube 79 string

Scenario requires quite some tuning

• Large DM density at Galactic center
• Cancellation of contributions to SI cross section to

avoid LUX constraints

• Exploitation of form factors for SD cross section to

avoid IceCube constraints

• Allow addition uncertainty in the modeling of the

gamma-ray signal to obtain agreement with excess emission Features:

• Point is very constrained
• Almost perfect agreement with observed relic density
• This scenario will be tested in the very near future in

various ways.

57

C h e c k l i s t

Unresolved PSCs (e.g. MSPs) Unresolved PSCs (e.g. MSPs) Problems in the subtraction of Problems in the subtraction of Galactic foreground / Bubbles Galactic foreground / Bubbles Recent burst-like events (protons) Recent burst-like events (protons) Recent burst-like events (electrons) Recent burst-like events (electrons) Dark matter annihilation Dark matter annihilation Spectrum Spectrum Morphology Morphology

? ? ? ? OK OK OK OK OK OK ? ? ? ? ? ?

58

Ma n y

• p

e n q u e s t i

• n

s

Milli-second pulsars

• Population studies and modeling
• Searches for unresolved gamma-ray point sources

[Fermi-LAT, Gamma-400, AstroGam, Pangu]

• X-ray & radio observations

Non-standard diffusion models

• Modeling of anisotropic diffusion, convective winds
• Searches for synchrotron emission & Bremsstrahlung

Corroborating evidence for dark matter annihilation

• Gamma-ray observations of dwarf spheroidals
• Radio observations of dwarf spheroidals [e.g. Regis+ 2014]
• Searches with antiprotons
• Direct searches & collider searches

59

O n e

• p
• i

n t f l u c t u a t i

• n

s e t c

[Lee, Lisanti, Safdi 2014] One-point fluctuations can help to discriminate just-below-threshold point sources from a diffuse dark matter-induced flux.

60

C

• n

c l u s i

• n

s

• We performed first comprehensive analysis of BG systematics for

the Fermi GeV excess in the inner Galaxy

• Theoretical model systematics: From 60 GDE models
• Empirical model systematics: From PCA of residuals
• We defined robust statistical tools to describe spectral and

morphological properties of the excess emission Results

• We robustly confirm the existence of the Fermi GeV excess in the

inner Galaxy

• The spectrum features a peak at 1-3 GeV and is best fit with a

broken power law. Excellent fits also with DM spectra possible.

• GeV excess extends to at least 10 degree away from GC at 95% CL
• Compatible with uniform spectrum and spherical symmetry within

95% CL

• This suggests: DM annihilation, unresolved point sources, maybe

leptonic burst event, ...

• Outlook: Multi-wavelength, multi-messenger, ...