Indirect Detection of Dark Matter with Gamma Rays

Indirect Detection of Dark Matter with Gamma Rays Simona Murgia SLAC-KIPAC University of California, Irvine on behalf of the Fermi LAT Collaboration Dark Attack 212 Ascona, Switzerland 15-2 July 212

WIMP Searches COLLIDER SEARCHES Fermilab LHC DM SM DIRECT SEARCHES CDMS? XENON DM INDIRECT SEARCHES Fermi-LAT SM PAMELA IceCube

WIMP Signal χ γ d de Gamma rays from DM annihilation: particle physics 1 < ann v > (E,, ) = 4 2m 2 W IMP d (, ) los f dn f de B f 2 (r(l, ))dl(r, ) Photon Flux [cm -2 s -1 GeV -1 ] -4-5 -6-7 -8-9 - -11-12 -13-14 1 χ?? Neutralino continuum gamma ray flux towards galactic centre - NFW model,!"= -3 sr γ, Z,... -2.7 Background, E # 3 GeV neutralino added 5 GeV neutralino added 2 3 4 5 6 2 3 4 5 6 2 3 4 5 6 Photon energy [GeV] Bergstrom, Ullio, Buckley DM distribution Bertone et al., arxiv:811.3744 u, d, s, c τ e μ W, Z, h, b, t MDM=1TeV Cirelli et al, arxiv:89.249

Dark Matter Distribution We generally heavily rely on simulations of the dark matter distribution to make predictions for DM searches...... but much is still unknown on how DM is distributed, e.g.: cuspiness of the profile halo shape (spherical, prolate, oblate, triaxial, dark disk,...) substructure Walker et al 29 Bertone et al., arxiv:811.3744

Fermi Mission The Large Area telescope The Fermi Large Area Telescope observes the gamma-ray sky in the 2 MeV to >3 GeV energy range with unprecedented sensitivity Large Area Telescope (LAT) Orbit: 565 km, 25.6 o inclination, circular. The LAT observes the entire sky every ~3 hrs (2 orbits) Gamma-ray Burst Monitor (GBM) 8 kev - 4 MeV

Fermi Mission The Large Area telescope The Fermi Large Area Telescope observes the gamma-ray sky in the 2 MeV to >3 GeV energy range with unprecedented sensitivity Orbit: 565 km, 25.6 o inclination, circular. The LAT observes the entire sky every ~3 hrs (2 orbits) Large Area Telescope ~1.8 m(lat) γ Tracker Gamma-ray Burst Monitor (GBM) 8 kev - 4 MeV ACD e - e + Calorimeter

Direct Measurement of Local Cosmic Rays Fermi LAT measured the cosmic ray e + +e - spectrum from 7 GeV to 1 TeV Fermi LAT confirms the raise in the cosmic ray positron fraction observed by PAMELA and extends it to 2 GeV Measurements consistent with a nearby source contributing to the observed e + and e - DM interpretation can be constrained by gamma-ray data (shown later) Fermi LAT Collaboration 2, Phys.Rev. D82 924 Fermi LAT Collaboration 212, Phys.Rev.Lett. 8 113

Understanding the Gamma-ray Sky Fermi LAT data - 3 years = + + + point sources Galactic diffuse isotropic (and Sun+Moon) dark matter??

Gamma Rays from Dark Matter Pieri et al 211, Phys.Rev.D 83 23518

Gamma Rays from Dark Matter Satellites: Spectral lines: Low background and good source ID, but low statistics JCAP 124 (212) 16 ApJ 747, 121 (212) Phys. Rev. Lett. 7, 24132 (212) ApJ 712, 147 (2) JCAP 1 (2) 31 ApJ 718, 899 (2) Good source ID, but low statistics Phys. Rev. D, In press (212) Phys. Rev. Lett. 4, 9132 (2) Galaxy clusters: Low background but low statistics JCAP 5 (2) 25 Galactic center: Good statistics but source confusion/diffuse background Milky Way halo: Large statistics but diffuse background arxiv:125.6474 +Electrons and Positrons! Phys.Rev.Lett., 8 113 (212) Phys. Rev. D84, 327 (211) Nucl. Instrum. Meth. A63 (211) 48-51 Phys. Rev. D82, 923 (2) Phys.Rev.Lett., 2 1811 (29) Anisotropies MNRAS 414, 24 (211) Extragalactic diffuse: Large statistics, but astrophysics, Galactic diffuse background JCAP 4 (2) 14 Pieri et al 211, Phys.Rev.D 83 23518

Galactic Diffuse Emission The diffuse gamma-ray emission from the Milky Way is produced by cosmic rays interacting with the interstellar gas and radiation field and carries important information on the acceleration, distribution, and propagation of cosmic rays. = + + data point sources galactic diffuse isotropic Inverse Compton Bremsstrahlung π -decay x-ray, gamma-ray x-ray, gamma-ray proton

Galactic Diffuse Emission Cosmic ray origin, propagation, and properties of the interstellar medium can be constrained by comparing the data to predictions Generate models (in agreement with CR data) varying CR source distribution, CR halo size, gas distribution (GALPROP) and compare with Fermi LAT data Example model: CR source distribution: SNRs CR confinement region: 2 kpc radius, 4 kpc height (data - prediction)/prediction) for example model Fermi LAT data, 21 months, 2 MeV to GeV Fermi LAT Collab. 212, ApJ 75, 3 DGE π -decay bremsstrahlung IC sources Total isotropic Inner galaxy

Galactic Diffuse Emission Cosmic ray origin, propagation, and properties of the interstellar medium can be constrained by comparing the data to predictions Generate models (in agreement with CR data) varying CR source distribution, CR halo size, gas distribution (GALPROP) and compare with Fermi LAT data Example model: CR source distribution: SNRs CR confinement region: 2 kpc radius, 4 kpc height On a large scale the agreement between data and prediction is overall good, however some extended excesses stand out. (data - prediction)/prediction) for example model Fermi LAT data, 21 months, 2 MeV to GeV Fermi LAT Collab. 212, ApJ 75, 3 DGE π -decay bremsstrahlung IC sources Total isotropic Inner galaxy

Extended Lobe-like Features in the Fermi Sky Gamma-ray bubbles (Su et al 2, ApJ 724, 44): very extended (~ 5 o from plane) hard spectrum (~E -2, 1- GeV) sharp edges possible counterparts in other wavelengths (ROSAT, WMAP, and Planck) Outflow from the center of the Milky Way: jets from the supermassive black hole? starburst? Fermi LAT data, E> GeV Su et al 2, ApJ 724, 44

Galactic Center Region Steep DM profiles predicted by CDM Large DM annihilation/decay signal from GC! Good understanding of the conventional astrophysical background is crucial to extract a potential DM signal from this complex region of the sky: sources: many energetic sources near to or in the line of sight of the GC Galactic diffuse emission modeling: large uncertainties complicated by overlap of structures along the line of sight, difficult to model. Also, unresolved source component likely. Large uncertainties in the dark matter distribution cored Bertone et al., arxiv:811.3744 Outside GC region GC region

Fermi LAT View of the Galactic Center Region Fermi LAT preliminary results for a 15 o x15 o region around the direction of the Galactic center with 32 months of data, 1- GeV. P7CLEAN data, FRONT only. Bright excesses after subtracting diffuse emission model are consistent with cataloged sources Fermi LAT Collaboration, paper in preparation W28 2FGL J1745.6-2858 LAT PSR J1732-3131 LAT PSR J189-2332 DATA DATA-MODEL (diffuse) zero suppressed Counts/.1 deg 2 Counts/.1 deg 2

Fermi LAT View of the Galactic Center Region Fermi LAT preliminary results for a 15 o x15 o region around the direction of the Galactic center with 32 months of data, 1- GeV. P7CLEAN data, FRONT only. Diffuse emission model and contribution from detected point sources account for most of the emission observed in the region Fermi LAT Collaboration, paper in preparation DATA DATA-MODEL (diffuse+sources) Counts/.1 deg 2 Counts/.1 deg 2

Fermi LAT View of the Galactic Center Region Fermi LAT preliminary results for a 15 o x15 o region around the direction of the Galactic center with 32 months of data, 1- GeV. P7CLEAN data, FRONT only. Diffuse emission model and contribution from detected point sources account for most of the emission observed in the region Good agreement between data and model within 5-%. Fermi Investigating LAT Collaboration, low paper level in preparation residuals Dark matter analysis ongoing DATA DATA-MODEL (diffuse+sources) Counts/.1 deg 2 Counts/.1 deg 2

Dwarf Spheroidal Galaxies Optically observed dwarf spheroidal galaxies (dsph): largest clumps predicted by N-body simulation. Excellent targets for gamma-ray DM searches Very large M/L ratio: to ~> (M/L ~ for Milky Way) DM density inferred from the stellar data! Data so far cannot discriminate, in most cases, between cusped or cored dark matter profiles. However, Fermi s DM constraints with dsph do not have a strong dependence on the inner profile ultra-faint dwarfs classical dwarfs Expected to be free from other gamma ray sources and have low dust/gas content, very few stars Only a handful of dwarfs are known (SDSS) but more should be discovered (DES, LSST)

Dwarf Spheroidal Galaxies Bootes I, Carina, Coma Berenices, Draco, Fornax, Sculptor, Segue I, Sextans, Ursa Major II, Ursa Minor No detection of dsph by Fermi with 2 years of data Determine 95% flux upper limits for several possible annihilation final states Combine with the DM density inferred from the stellar data (assume NFW profile) to set constraints on the annihilation cross section Constraints include systematic uncertainties on the DM content! Thermal WIMPs Fermi LAT Collaboration, Phys. Rev. Lett., 7 (211) 24132 17

Dwarf Spheroidal Galaxies Bootes I, Carina, Coma Berenices, Draco, Fornax, Sculptor, Segue I, Sextans, Ursa Major II, Ursa Minor No detection of dsph by Fermi with 2 years of data Determine 95% flux upper limits for several possible annihilation final states Combine with the DM density inferred from the stellar data (assume NFW profile) to set Fermi constraints LAT rules on the out annihilation some WIMP cross models section with generic cross section (3x -26 cm 3 s -1, s-wave) for a Constraints thermal relic include systematic uncertainties on the DM content! Thermal WIMPs Fermi LAT Collaboration, Phys. Rev. Lett., 7 (211) 24132 17

Cotta, R. C. et al., JCAP 124 (212) 16 Constraining the pmssm with Fermi LAT Use combined likelihood formalism with dsphs to constrain the phenomenological MSSM (19 parameter scan, ~7k models survive existing experimental bounds Closest model to discovery/exclusion ~1.5x away Assuming an improvement in sensitivity of ~x with a year mission, many models within reach Complementarity with direct detection experiments Blue: models that saturate relic density Grey: all models Fermi LAT yrs XENON LUX (projected) SuperCDMS (projected) COUPP 6, 5 kg (projected)

Search for Unknown Dark Matter Satellites Search the Fermi sky for DM satellites (1 year of data, 2 MeV-3 GeV) DM satellite candidates ( b >2 o ): unassociated sources in 1FGL (231) non-1fgl source candidates (154) Test for: DM spectrum (power law vs b-bbar) Test for spatial extension (point vs NFW) Two sources pass tests, but rejected after further inspection 1FGL J3+451 No DM satellites found Set 95% CL upper limit for GeV WIMP annihilating into b-bbar, 1.95 24 cm 3 s 1 Fermi LAT Collaboration, Astrophys. J. 747 (212) 121 19

Fermi LAT Collaboration, Phys. Rev. D (212), in press Search for Spectral Lines Smoking gun signal of dark matter. χ γ The line signal is generally suppressed (but enhanced in some models!)?? χ γ, Z, H,...... The signal is the LAT line response function. The background is modeled by a power-law function and determined by the fit Search for lines in the first 2 years of Fermi data and include the data from most of the sky (remove point sources and most of the Galactic disk) Fermi LAT data (5-264 GeV)

Fermi LAT Collaboration, Phys. Rev. D (212), in press Search for Spectral Lines No line detection. 95% CL flux upper limits, 7-2 GeV energy range With assumptions on the dark matter density distribution, we extract constraints on the dark matter annihilation cross-section and decay lifetime

Fermi LAT Collaboration, Phys. Rev. D (212), in press Search for Spectral Lines No line detection. 95% CL flux upper limits, 7-2 GeV energy range Limits constrain lifetime for gravitino decay Ibarra, A et al, Phys. Rev. Lett., 6131 (28) With assumptions on the dark matter density distribution, we extract constraints on the dark matter annihilation cross-section and decay lifetime Wino LSP (*) Thermal MSSM WIMP lines (*) Acharya, B et al, JHEP 15 (211) 33

More DM constraints from Lines Constraints have also been placed on recently-proposed models that predict WIMPs annihilating into γ+higgs. Higgs in space! Jackson, C et al JCAP 4 (2) 4 E 2 dfêde @GeV cm -2 s -1 D -4-5 -6-7 -8-9 - -11-12 -13-14 -15-16 M n =149 GeV Hg Z' n =g Z' t =3L M n =162 GeV Hg Z' n =g Z' t =1L g Z - M Z' =22 GeV g h- M h =125 GeV g Z EGRET FERMI HESS 1 2 E g @GeVD

Inclusive spectrum limits Fermi LAT Collaboration, Phys. Rev. D (212), in press

Inclusive spectrum limits Fermi LAT Collaboration, Phys. Rev. D (212), in press Strongly disfavors DM interpretation of Fermi/PAMELA electron and positron measurements for annihilation into taus and constrains annihilation into muons

A spectral Line from the Galactic Center Region? Fermi < E < 12 GeV 2.5 9 9 2.5 Edmonds, PhD Thesis 211 kev cm-2 s-1 sr-1 2. 2. 45 45 1.5 1.5 1. 1..5-45. -9 18 9.5-45. -9-18 -9 18 9-9 -18 Fermi 14 < E < 16 GeV 2.5 9 9 2.5 2. 45 45 1.5 1.5 1. 1..5-45.5-45 Weniger, arxiv:124.2797.-9. 9-9 -18-9 18-9 -18 Residual map Weniger, arxiv:124.2797 2.5 9 9 Su et al, arxiv:126.1616 Fermi 16 < E < 18 GeV 9 2.5 2. 2. 45 45 1.5 1.5 1. 1..5.5-45. -9 18. -45 12-14 GeV residual map Optimized line search region 9-9 -18-9 18 -.5-1. 9-9 -18 kev cm-2 s-1 sr-1 kev cm-2 s-1 sr-1 kev cm-2 s-1 sr-1 kev cm-2 s-1 sr-1 2. 18 kev cm-2 s-1 sr-1 Fermi 8 < E < GeV Recent papers on evidence for a line at 13 GeV (Bringmann et al, arxiv:123.1312 (internal brem), Weniger, arxiv:124.2797, Su et al, arxiv:126.1616) LAT collaboration has published line search with upper limits using 2 years of data (N.B. Different studies have used different search regions) developed control samples to evaluate instrumental effects (e.g. Earth limb) Fermi 12 < E < 14 GeV Comprehensive Fermi LAT team analysis on line searches based on 4 years of data (Pass 7) ongoing

A spectral Line from the Galactic Center Region? Fermi < E < 12 GeV 2.5 9 9 Claimed 13 GeV dark matter line 45 (C. Weniger, arxiv:124.2797) 2. kev cm-2 s-1 sr-1 2.5 Edmonds, PhD Thesis 211 45 2. 1.5 1.5 1. 1..5-45. -9 18 9.5-45. -9-18 -9 18 9-9 -18 Fermi 14 < E < 16 GeV 2.5 9 9 2.5 2. 45 45 1.5 1.5 1. 1..5-45.5-45 Weniger, arxiv:124.2797.-9. 9-9 -18-9 18-9 -18 Residual map Weniger, arxiv:124.2797 2.5 9 9 Su et al, arxiv:126.1616 Fermi 16 < E < 18 GeV 9 2.5 2. 2. 45 45 1.5 1.5 1. 1..5.5-45. -9 18. -45 12-14 GeV residual map Optimized line search region 9-9 -18-9 18 -.5-1. 9-9 -18 kev cm-2 s-1 sr-1 kev cm-2 s-1 sr-1 kev cm-2 s-1 sr-1 kev cm-2 s-1 sr-1 2. 18 kev cm-2 s-1 sr-1 Fermi 8 < E < GeV Recent papers on evidence for a line at 13 GeV (Bringmann et al, arxiv:123.1312 (internal brem), Weniger, arxiv:124.2797, Su et al, arxiv:126.1616) LAT collaboration has published line search with upper limits using 2 years of data (N.B. Different studies have used different search regions) developed control samples to evaluate instrumental effects (e.g. Earth limb) Fermi 12 < E < 14 GeV Comprehensive Fermi LAT team analysis on line searches based on 4 years of data (Pass 7) ongoing

A spectral Line from the Galactic Center Region? Astrophysical interpretations have been proposed, e.g.: Spectral and spatial variations of the diffuse gamma-ray background in the vicinity of the Galactic plane and possible nature of the feature at 13 GeV Boyarsky et al, arxiv:125.47 Cold ultrarelativistic pulsar winds as potential sources of galactic gamma-ray lines above GeV Aharonian et al, arxiv127.458 Possible line feature in Earth limb data (C. Weniger, talk at Gamma212, similar analysis also in Su et al) Plots from C. Weniger talk at Gamma212

Very High Energy Gamma rays MAGIC Imaging Atmospheric Cherenkov Telescopes (IACTs) VERITAS H.E.S.S. Sky coverage and sources VERITAS/MAGIC http://tevcat.uchicago.edu H.E.S.S.

Ground vs Space Gamma-ray experiments Lower energy thresholds accessible in space, and up to ~ TeV energies with experiments on the ground. Overlap in the ~ GeV region Larger field of view, great duty cycle, and all sky coverage in space Single photon angular resolution: ~1 o at 1 GeV (Fermi LAT), ~.1 o at GeV (ACTs, Fermi LAT), ~.5 o at 1 TeV (ACTs) Energy resolution: ~ 8% at GeV (Fermi LAT), ~15% at 1 TeV (ACTs) Large collecting area on the ground (high sensitivity) Fermi LAT FoV ACT FoV Plot from S. Funk Fermi LAT Point Spread Function

H.E.S.S.: Galactic Halo GC is complicated by astrophysics, look away from it! Signal region: relatively close to GC but free from astrophysical background Select a region where the contribution from DM is smaller for background subtraction (background region) Small dependence on DM profile Abramowski et. al. Phys.Rev.Lett.6 (arxiv:13.3266)

Indirect detection with γ rays: Current Best constraints J. Conrad, talk at Gamma212 <σv> ( -26 cm 3 s -1 ) s - Current 15

Search for DM in Galaxy Clusters Most massive structures observed in the Universe, very large dark matter content. Predicted to emit gamma rays from conventional astrophysical processes Search the Fermi sky for galaxy clusters (2 years of data, 2 MeV- GeV), modeled as point sources No detection of galaxy clusters Set constraints on annihilation cross section Claim of extended emission from the Virgo galaxy cluster from Han et al (arxiv:121.3). However foreground due to local extended structure (Loop I) complicates interpretation. Fermi LAT Collaboration, paper in preparation Thermal WIMP

Fermi LAT Collaboration, JCAP 4 (2) 14 Extragalactic Gamma-ray Background Fermi LAT preliminary results for 24 months of data, 2 MeV-58 GeV, P7ULTRACLEAN data data - - = point sources Galactic diffuse DM constrains for annihilation into b quarks Large uncertainties on DM distribution PRELIMINARY

Summary/Outlook Very promising constraints on the nature of DM have been placed (e.g. dsph, lines) The astrophysical background is currently a limitation in particular for the Galactic center, which has huge potential in terms of discovery or setting constrains Looking forward: Updated searches ongoing. Fermi LAT team paper on line searches, based on 4 years of data upcoming Fermi continues to survey the sky (NASA Senior Review recommended extending operations through 216, at least). Fermi gamma-ray data are public immediately! Improvements in Fermi LAT event analysis will crucially improve the reach of some of the searches, e.g. improved PSF will somewhat mitigate source confusion in the GC region Some analyses will also greatly benefit from multi-wavelength observations, e.g. dsph, GC Future experiments (e.g. CTA) will significantly improve the energy reach of indirect dark matter searches with gamma rays

Anisotropies Fermi can test the nearby source hypothesis looking for anisotropies in the e + + e - sky No significant anisotropies were found in Fermi electron data (angular scales from o to 9 o ) However upper limits on dipole anisotropy cannot yet rule out individual pulsar/dm interpretation of PAMELA and Fermi e + e - data Pre-trial significance map ( o, E>6 GeV) Fermi LAT Collaboration, Phys.Rev. D82 (2) 923 Fermi UL Vela GALPROP Monogem 33

-26 2 Galactic Halo cc ô bb, NFW -22 IC+FSR, wêo background wêo background modeling modeling FSR, wêo free background modeling constrained source fits IC+FSR, constrained free source fits 3s 5s -22-22 -23-23 cc ô bb, NFW IC+FSR, wêo background modeling FSR, wêo background 3s, wêo background modeling, rmodeling =.2-.7 GeV cm-3 IC+FSR, constrained free source fits wêo background modeling constrained 3s free source fits 3s, r5s =.43 GeV cm-3 5s, r=.43 GeV cm-3-24 -24-25 -22-23 -24-25 swimp freeze-out -25 Preliminary 2 m @GeVD 3 cc ô m m, NFW m @GeVD 2 + - 4 3 <sv> @cm3s -1D -24-24 -25-25 swimp freeze-out -26-26 -26-21 4 2-26 -26 647 m @GeVD 34.1 5-4 1-4 5-5.1 2 2 m @GeVD m @GeVD 3 16 + cc ôcc m+ô m, tiso t, NFW cc ô bb, ISO -21-21 IC+FSR, wêo background modeling IC+FSR, wêo background modeling FSR, wêo background modeling FSR, wêo background modeling wêo background modeling IC+FSR, constrained 3s free source fits fits free source -22source fits IC+FSR, constrained constrained 5s -22 free 3s 3s 5s 5s -23-23 <sv> @cm3s -1D -21.5 swimp freeze-out swimp freeze-out Fermi LAT Collaboration, arxiv:125.6474 <sv> @cm3s -1D -21 <sv> @cm3s -1D Limit the analysis to regions of the galactic halo better -23-23 described by the model. -23 Robust limits can be determined by requiring that -24 a DM -24-24 contribution does not over-predict the data in these regions Data: 2 years, 1- GeV (w/ background modeling), -25-25 1-4 -25 swimp freeze-out GeV (w/o background modeling) -26-26 Constrains DM interpretation of Fermi/PAMELA electron -26 2 3 4 and positron measurements. Significant improvements m @GeVD forseen as modeling of the diffuse background improves 3s 5s <sv> @cm3s -1D -21 3 4 4.5.1 5-4 1-4 5-5.1 <sv> @cm3s -1D -22 cc ô m m, NFW E 2F @MeV cm-2 s -1 sr-1d -21 E 2F @MeV cm-2 s -1 sr-1d 3s, wêo background modeling, r=.2-.7 GeV cm wêo background modeling constrained free source fits 3s, r=.43 GeV cm-3 5s, r=.43 GeV cm-3 cc ô m+ m-, NFW 4 16 cc ô bb, ISO + - -3-22 m @GeVD 3 FIG. 7: Counts map of the ROI that we consider (upper left panel), mod 2 best fit (zh = kpc, e,2 = 2.3 and d2hi=.14 2 mag cm ) param.5 for the same model (second row left) and when DM (of mass m =15 Ge (second row right). Third row: same as second row but with 2FGL point swimp mask (left) andfreeze-out 2FGL mask (right). The model and data counts.1 and the -4 5 filter. The point sources mask in the residuals have been applied before a swimp s freeze-out WIMP freeze-out Preliminary E 2F @MeV cm-2 s -1 sr-1d -21 <sv> @cm3s -1D <sv> @cm3s -1D 3-1 <sv> @cm <sv> @cm3ss -1DD swimp freeze-out 3 4 1-4 fit of our analysis, was chosen for the reference model: va5 =36 km -5 2 2 = 2.39, 3 br,p = 11.5, 3 = 2.45, d2hi=.14 648 p,1 = 1.9,p,2 4 4 2 e,2 649 are shown in Table III..1 cc ô m+ m-, ISO m @GeVD m @GeVD

Dwarf Spheroidal Galaxies Bootes I, Carina, Coma Berenices, Draco, Fornax, Sculptor, Segue I, Sextans, Ursa Major II, Ursa Minor No detection of dsph by Fermi with 2 years of data Determine 95% flux upper limits for several possible annihilation final states Combine with the DM density inferred from the stellar data (assume NFW profile) to set constraints on the annihilation cross section Constraints include systematic uncertainties on the DM content! L(D p W,{p} i )= Y i L LAT i (D p W, p i ) 1 ln() J i p 2 i e ( log (J i) log (J i)) 2 /2 2 i