Dark matter gamma-ray line searches toward the Galactic Center halo - PowerPoint PPT Presentation

Dark matter gamma-ray line searches toward the Galactic Center halo with H.E.S.S. I Emmanuel Moulin, Lucia Rinchiuso, Aion Viana, Christopher van Eldik, Johannes Veh for the H.E.S.S. collaboration

The H.E.S.S. experiment § H.E.S.S. phase I § H.E.S.S. phase II § four 12m telescopes § four 12m telescopes § FoV 5 deg § one 28m telescope (FoV 3.5 deg) § energy threshold 100 GeV § energy threshold O(30 GeV) § angular resolution < 0.1 deg § angular resolution from 0.4 deg to less than 0.1 deg § Operating since 2003 H.E.S.S. phase I H.E.S.S. phase II Emmanuel Moulin . ICRC 2017, Busan, Korea 2

Why VHE gamma-rays ? § Do not suffer from propagation effects at Galactic Scale : they point back to the source § Can reveal the abundance and distribution of DM § Characteristic features may be present in the spectrum at these energies Emmanuel Moulin . ICRC 2017, Busan, Korea 3

Why VHE gamma-rays ? § Do not suffer from propagation effects at Galactic scale : they point back to the source § Can reveal the abundance and distribution of DM § Characteristic features may be present in the spectrum at these energies § Identification of DM is possible - the gamma-ray distribution in the sky tells the DM density distribution - the gamma-ray spectrum tells the reaction process and DM mass Emmanuel Moulin . ICRC 2017, Busan, Korea 4

Dark matter search with VHE gamma rays Particle Physics : Astrophysics Cross sections o → DM distribution in the target Differential photon yield o DM particle mass o Emmanuel Moulin . ICRC 2017, Busan, Korea 5

Dark matter search with VHE gamma rays Particle Physics : Astrophysics Cross sections o → DM distribution in the target Differential photon yield o DM particle mass o § Weakly Interacting Massive Particles (WIMPs) – The weak interaction mass scale and ordinary gauge couplings give right relic DM density without fine-tuning. – Mass scale GeV-TeV, makes them Cold Dark Matter – Provides benchmark for indirect detection: thermally-produced WIMPs Emmanuel Moulin . ICRC 2017, Busan, Korea 6

Spectral signatures in gamma rays Continuum emission (“Secondary photons”) → from fragmentation of quarks/massive gauge bosons (via π 0 decay) Gamma-ray lines → from two-body annihilation into photons → forbidden at tree-level, generically suppressed by O(α 2 ) Virtual Internal Bremsstrahlung (VIB) → radiative correction to processes with charged final states → generically suppressed by O(α) Emmanuel Moulin . ICRC 2017, Busan, Korea 7

Energy spectrum from dark matter annihilation 1. Continuum: hadronization and/or decay of W/Z, quarks, leptons… 2. Line from prompt annihilation in two photons not at tree level: suppressed but clear signature at DM mass ! 3. Final state radiation 4. Virtual internal bremsstrahlung “power-law” background The spectral information can be γ- line w/o exploited for a better energy signal vs. resolution background discrimination cut-off of the signal at DM mass Emmanuel Moulin . ICRC 2017, Busan, Korea 8

Dark matter targets gamma-ray searches Galaxy satellites of the Milky Way Substructures in Many of them within the 100 kpc from GC o the Galactic halo High M/L o Lower signal o Low astrophysical background o Cleaner signal o (once found) Galactic Centre Proximity (~8kpc) o Possibly high DM o Galactic halo concentration : Large statistics o DM profile : core? cusp? Galactic diffuse o High astrophysical o background bck / source confusion Aquarius, Springel et al. Nature 2008 § DM density matters § Astrophysical background matters as well Emmanuel Moulin . ICRC 2017, Busan, Korea 9

Dark matter distribution in the Galactic Centre region (1) We assume cusped DM density profiles, e.g. the Einasto, NFW, … per bin of 0.02degx0.02deg Einasto profile ρ $%&' r = ρ * exp −2 r 2 − 1 α r * parametrized with α = 0.17 r * = 21 𝑙𝑞𝑑 ρ * = 0.07 𝐻𝑓𝑊𝑑𝑛 >? as used in HESS GC 2011, 2013 papers Emmanuel Moulin . ICRC 2017, Busan, Korea 10

Dark matter distribution in the GC region (2) We assume cusped DM density profiles, e.g. the Einasto, NFW, … Einasto profile per bin of 0.02degx0.02deg Spatial information can be used to discriminate from the isotropic residual background Emmanuel Moulin . ICRC 2017, Busan, Korea 11

Dark matter distribution in the GC region (3) We assume cusped DM density profiles, e.g. the Einasto, NFW, … Einasto profile per bin of 0.02degx0.02deg Example for one OFF region for a given pointing ON region pointing position § Significant DM gradient between ON and OFF regions Emmanuel Moulin . ICRC 2017, Busan, Korea 12

Optimized background measurement § OFF regions are chosen symmetrically to ON regions with respect to the pointing position in the same observational field of view § Overlapping areas are discarded to maintain the dark matter gradient → same acceptance in ON and OFF region (due to azimuthal symmetry) → strong dark matter gradient that improves the limits Emmanuel Moulin . ICRC 2017, Busan, Korea 13

The Galactic Centre region seen by H.E.S.S. I § Largest H.E.S.S. dataset from observations accumulated over 10 years § Most detailed view of GC @VHE : pointlike sources (PWN, SNR, Unid) and diffuse emission → GC region is a crowded environment at VHE Excess map, E>200 GeV 75 pc § DM density in the central region of the Milky Way expected to be large Emmanuel Moulin . H.E.S.S. Moriond EW 2015 14

Region of interest (ROI) definition Galactic Longitude Gamma-ray excess map G0.9 Sgr A* Galactic Latitude HESS J1745-303 § 10-year observations with H.E.S.S. 1 toward the GC § Gamma-ray statistics : 254 h compared to 112 h (2011/2013 papers) § Novel analysis method : 2D likelihood analysis with spectral and spatial information of signal and background Emmanuel Moulin . ICRC 2017, Busan, Korea 15

Region of interest (ROI) definition § Excluded region ± 0.3 ° : dominated by astrophysical sources and diffuse emission § Whole region of Interest: 1 ° in radius § ROI divided in 7 sub-regions of 0.1 ° : spatial binning Emmanuel Moulin . ICRC 2017, Busan, Korea 16

� � � Gamma-ray line rate 𝛅 = 𝟐 𝛕𝐰 [ 𝐞𝐎 𝐞𝐅 O 𝐒 𝐅, 𝐅 O 𝐔 𝐩𝐜𝐭 𝐁 𝐟𝐠𝐠 (𝐅)𝐞𝐅′ 𝐎 𝐣𝐤 x 𝐊(∆𝛁) 𝟑 𝟓𝛒 𝟑𝐧 𝐄𝐍 § Spectrum : Dirac delta function at 𝐧 𝐄𝐍 𝐞𝐎 𝐞𝐅′ = 𝟑𝛆 𝐅 O − 𝐧 𝐄𝐍 > 𝐅>𝐅 U 𝟑 𝟐 𝐒 𝐅, 𝐅 O = 𝐟 𝟑𝛕 𝟑 § Energy resolution 𝟑𝛒 𝛕 𝛕 𝐅 = 𝟐𝟏% with sigma/E - almost independent of the offset and the azimuth angle bias Emmanuel Moulin . ICRC 2017, Busan, Korea 17

2D-binned likelihood analysis § Likelihood function binned in energy (bin j) and space (RoI, bin i) O + 𝐎 𝐂 𝐎 𝐏𝐆𝐆 𝓜 𝐣𝐤 𝐎 𝐏𝐎 , 𝐎 𝐏𝐆𝐆 𝐎 𝐓 , 𝐎 𝐂 = 𝐎 𝐓 + 𝐎 𝐂 𝐎 𝐏𝐎 𝐎 𝐓 𝐟 > 𝐎 𝐂 r𝐎 𝐓U 𝐟 > 𝐎 𝐓 r𝐎 𝐂 𝐎 𝐏𝐎 ! 𝐎 𝐏𝐆𝐆 ! where expected N B obtained from dL/dNB = 0 𝐎 𝐏𝐎 Observed signal (ON region) 𝐎 𝐏𝐆𝐆 Observed background (OFF region) 𝓜 = h 𝓜 𝐣𝐤 𝐎 𝐓 Line signal (expected in ON) 𝐣,𝐤 𝐎′ 𝐓 Line signal (expected in OFF) 𝐎 𝐂 Expected background § Test statistics: likelihood-ratio test O , 𝐎 𝐂 𝓜 𝐎 𝐏𝐎 , 𝐎 𝐏𝐆𝐆 𝐎 𝐓 , 𝐎 𝐓 𝐔𝐓 = −𝟑 𝐦𝐩𝐡 𝓜 𝐗 = −𝟑 𝐦𝐩𝐡 𝓜 𝐗𝐏 𝓜 𝐎 𝐏𝐎 , 𝐎 𝐏𝐆𝐆 𝐎 𝐂 TS = 2.71 § No significant excess if found in any of the ROIs : for a 95% CL limit Emmanuel Moulin . ICRC 2017, Busan, Korea 18

95% C.L. expected limits on the annihilation flux § Improvement of a factor about 8 observed @ 1 TeV on the mean expected limits § Improvement from the analysis: exclusion regions, 2D binned likelihood analysis approach, improved raw data danalysis, higher statistics Emmanuel Moulin . ICRC 2017, Busan, Korea 19

95% C.L. expected limits on the annililation cross section σv § best limit 2x10 -28 cm 3 s -1 @1TeV § mass range extended down to 300 GeV and up to 70 TeV § lower energy threshold thanks to the improved raw data analysis: best limit shifted down to lower masses Emmanuel Moulin . ICRC 2017, Busan, Korea 20

Comparison with previous result and other experiments § best limit 2x10 -28 cm 3 s -1 @1TeV § lower energy threshold thanks to the improved raw data analysis: best limit shifted down to lower masses § Fermi-LAT limits surpassed of a factor about 6 @300 GeV Emmanuel Moulin . ICRC 2017, Busan, Korea 21

Summary § Full H.E.S.S.-I GC dataset analysis for dark matter line searches in the inner GC halo in the DM mass range 300 GeV – 70 TeV § Improvement at 1 TeV with respect to 2013 results is about 8 on the mean expected sensitivity - higher photon statistics - higher sensitivity of the improved raw data analysis - 2D likelihood analyis method § Next steps : - Observed limits for publication - Follow-up study with H.E.S.S.-II data: more statistics and extended pointings from inner Galaxy survey started in 2015 - Tests of different TeV DM models can be performed Emmanuel Moulin . ICRC 2017, Busan, Korea 22