Koji Ichikawa. Prospects on the indirect dark matter detection and a future spectroscopic survey of dwarf spheroidal galaxies.

Transcription

1 Prospects on the indirect dark matter detection and a future spectroscopic survey of dwarf spheroidal galaxies Koji Ichikawa (In preparation) In collaboration with Kohei Hayashi, Masahiro Ibe, Miho N. Ishigaki, Shigeki Matsumoto and Hajime Sugai. TeVPa2015, Kashiwa, Oct ,

2 Direct Detection Dark Matter Search Indirect Detection DM SM DM SM Collider Production 2

3 Signal Target Milky-Way Galaxy dsphs 100 kpc 10 Mpc Extra Galaxy/ Cluster 8.5 kpc Charged CRs 3

4 dsphs: Dwarf spheroidal galaxies = Clean & DM Rich Target 1. Neighbor galaxies: 10~100kpc 2. Large Mass to Luminosity ratio = DM rich 3. Clean (no strong gamma-ray source) arxiv: v6 [astro-ph.co]

5 Current(slightly old) obs limit (ex. Wino) Collider Wino DM annihilation cross section 5

6 Current(slightly old) obs limit (ex. Wino) Collider Wino DM annihilation cross section 6

7 We Should Precisely Determine The dsph DM Halo Shape Dwarf galaxy γ-rays Observed γ-ray Flux DM Property Halo Profile (J-factor)

8 J-Factor DM Density profile ( r / r s s ) (1 r / r s 1 s (1 r / r ) 1 s ) (1 r / r 2 s ) 2 Cusp Cored Stellar Density Profile: ν(r) Jeans equation for stars 2 (Theory) l.o.s Fit 2 (obs) l.o.s Geringer-Sameth et al., arxiv:

9 Astrophysical Factor DM Density profile ( r / r s s ) (1 r / r s 1 s (1 r / r ) 1 s ) (1 r / r 2 s ) 2 Cusp Cored Stellar Density Profile: ν(r) Jeans equation for stars 2 (Theory) l.o.s 2 (obs) l.o.s Fit Classical: Well-determined Ultra-faint: Not well-determined. Prior dependence Factor 1.6 ~ 2 unc. :Conservative? 9

10 Hidden Systematics Non Spherical? => uncertainty Axisymmetric: Hayasi and Chiba., arxiv: Foreground Contamination? N < 100: O(1) uncertainty N 1000: < 0.4 By K. Hayashi-san (Preliminary) Prior Bias?/Cut? N < 100: > O(1) uncertainty Bonnivard et al., arxiv:

11 Hidden Systematics Non Spherical? => uncertainty Axisymmetric: Hayasi and Chiba., arxiv: Foreground Contamination? N < 100: O(1) uncertainty N 1000: < 0.4 Bonnivard et al., arxiv: < NMem < 100 Prior Bias?/Cut? N < 100: > O(1) uncertainty Bonnivard et al., arxiv: Nmem 1000

12 Hidden Systematics Non Spherical? => uncertainty Axisymmetric: Hayasi and Chiba., arxiv: Foreground Contamination? N < 100: O(1) uncertainty N 1000: < 0.4 Draco Segue1 Prior Bias?/Cut? N < 100: > O(1) uncertainty Martinez et al., arxiv:

13 Hidden Systematics Non Spherical? => uncertainty Axisymmetric: Hayasi and Chiba., arxiv: Foreground Contamination? N < 100: O(1) uncertainty N 1000: < 0.4 Prior Bias?/Cut? (For Ultra faint dsphs) N < 100: > O(1) uncertainty How to Reduce Them? -> Increase #NMem!

14 Prime Focus Spectrograph FoV 1.3 deg (diam) with 2394 Fiber MMFS (M. G. Walker et al,. (2007)) 14

15 Prime Focus Spectrograph FoV 1.3 deg (diam) with 2394 Fiber MMFS (M. G. Walker et al,. (2007)) 15

16 #Obs Star (<V) #Obs Star (<V) by M. Ishigaki Prime Focus Spectrograph More accurate DM profile estimation More Robust constraints

17 #Obs Star (<V) #Obs Star (<V) by M. Ishigaki Prime Focus Spectrograph More accurate DM profile estimation More Robust constraints

18 Strategy 1. Mock Observable: dsph Stellar + Foreground dsph Stellar Mock Boltzmann Equation under DM profile Foreground Mock Besancon Model (Robin+ (2003)) Member Star Mock Preliminary 2. Detector Convolution: 1. fix: dv = 3.0km/s 3. Fit: by (v, r) probability density. (unbinned) Foreground Mock Preliminary Obs 18

19 Strategy Fit without Foreground Star 1. Mock Observable: dsph Stellar + Foreground dsph Stellar Mock Boltzmann Equation under DM profile Foreground Mock Besancon Model (Robin+ (2003)) Member Star Mock Preliminary 2. Detector Convolution: 1. fix: dv = 3.0km/s 3. Fit: by (v, r) probability density. (unbinned) Foreground Mock Preliminary Obs 19

20 Foreground Contamination Outer Region = FG dominant Preliminary How to reduce the FG? => Cut ~ 10 % Contamination How to include the FG effect? => Model the FG dist. 30 < NMem < Bonnivard et al., arxiv:

21 Cut Strategy ROI Cut: 1.3 deg radius for 4 pointing Color Magnitude Cut Gravity Cut velocity Cut Teff, Chemical Cut do not so efficient Current (i ~ 20.) Member 420 FG 30 (w/o vcut: 130) PFS (i < 21) Member 700 FG 80 (w/o vcut: 550) This contamination is ignored = Biases dlogj~ 0.1 Member FG Chem. -> degenerate PFS (i < 21.5) Member 900 FG 100 (w/o vcut: 650)

22 Fit including FG model Member Fraction Prob. Dist. Of FG Member Parameter = halo information FG Parameter Can be considered to be Gaussian after several cuts. 22

23 FG func Parameters Outer Region Stars (Data w/o v cut can be used) => FG func parameter inc. its errors can be determined (errors => prior) FG Dominate dsph Dominate 1.3 deg radisu (4 pointings) ROI Preliminary Large ROI =Large FG sample = Small error of ffg params

24 Results Current Data (inc. FG contami) dlog(j) ~ 0.25? PFS (inc. FG contami) dlog(j) < 0.15? Preliminary Sum FG Mem Optimization is on-going

25 Summary Indirect detection is essential for DM search. Gamma-ray observation of dsph can give robust constraints on the DM annihilation cross section. Investigation of stellar kinematics (PFS) will play a crucial role. Reduction of foreground stars is important 25

26 Thank You! Koji Ichikawa In collaboration with Kohei Hayashi, Masahiro Ibe, Miho N. Ishigaki, Shigeki Matsumoto and Hajime Sugai. TeVPa2015, Kashiwa, Oct , 2015