1

Fermi_2011Mar.ppt

T. Mizuno et al.

Fermi-LAT Study of Cosmic-rays/Diffuse Gamma-rays and

Implications on Particle Physics

Mar. 7, 2011 @ Kyoto Univ.Tsunefumi Mizuno (Hiroshima Univ.)On behalf of the Fermi-LAT collaboration

2

Fermi_2011Mar.ppt

T. Mizuno et al.

Fermi 衛星による広がったガンマ線・宇宙線電子の観測と基礎物理への制限

2011 年 3 月 7 日 @ 京都大学基礎物理研究所

水野　恒史 ( 広島大学理学部 )On behalf of the Fermi-LAT collaboration

3

Fermi_2011Mar.ppt

T. Mizuno et al.

OutlineIntroductionFermi LAT instrumentationGalactic Diffuse Gamma-raysBehind the diffuse gs: EGB and DM searchCosmic-ray Electrons

4

Fermi_2011Mar.ppt

T. Mizuno et al.

Gamma-ray Sky

• GeV gamma-ray sky= Point sources + Diffuse Gamma-rays

Fermi-LAT 1 year all-sky map

>=80% of g-rays

3c454.3

Vela

Crab

Geminga

5

Fermi_2011Mar.ppt

T. Mizuno et al.

What makes Diffuse g-rays?

• Diffuse Gamma-rays= Cosmic-rays x (ISM, ISRF)

Planck microwave map = ISM gas

InterStellar MediumInterStellar Radiation Field

Galactic plane

nearby gas in high latitude

6

Fermi_2011Mar.ppt

T. Mizuno et al.

Gamma-ray Sky

• GeV gamma-ray sky~ Diffuse Gamma-rays = CRs x (ISM, ISRF)

Fermi-LAT 1 year all-sky map

phys. processes are well understood

7

Fermi_2011Mar.ppt

T. Mizuno et al.

Why are they important?

• Diffuse Gamma-rays are– “probe” to study Galactic CRs and ISM– “foreground” to study exotic physics, e.g.,

• signal from dark matter (DM)• extragalactic g-ray background (EGB)

Fermi-LAT 1 year all-sky map

new source classes or DM signal

annihilation or decay

8

Fermi_2011Mar.ppt

T. Mizuno et al.

Example: GeV Excess (EGRET Era)

• EGRET (1991-2000) reported excess emission when compared with a standard diffuse g-ray model– variety of explanations including DM signal

Hunter+97

G ~ 2.7(CR protons)

E2 x F

lux

0.1 1 10 GeV

excess

9

Fermi_2011Mar.ppt

T. Mizuno et al.

Example: GeV Excess (Fermi Era)

• EGRET GeV excess not confirmed.• However, Fermi data allow us investigating

more subtle “anomalies” (and detailed study of CRs/ISM)

EGRETFermi-LAT

G ~ 2.7Abdo+09PRL 103, 251101

E2 x F

lux

0.1 1 10 GeV

10

Fermi_2011Mar.ppt

T. Mizuno et al.

Fermi-LAT as Electron Detector

• Fermi-LAT reported a harder CR e- + e+ spectrum compared with a conventional model

Abdo+09PRL 102, 181101

Fermi Data (2009)

E3 x F

lux

10 100 1000 GeV

highest statistics:4.5 M events(6 months)

11

Fermi_2011Mar.ppt

T. Mizuno et al.

Fermi-LAT as Electron Detector

• Fermi-LAT reported a harder CR e- + e+ spectrum compared with a conventional model– Lots of interpretations (astrophysical and exotic)

Fermi e- + e+ PAMELA e+ ratio

+DM?Nearby objects?=

12

Fermi_2011Mar.ppt

T. Mizuno et al.

Fermi-LAT as Electron Detector

• Fermi-LAT reported a harder CR e- + e+ spectrum compared with a conventional model– Lots of interpretations (astrophysical and exotic)

2009 Jan. – 2010 Sep.

… and the paper is highly cited

13

Fermi_2011Mar.ppt

T. Mizuno et al.

Fermi-LAT Instruments

14

Fermi_2011Mar.ppt

T. Mizuno et al.

Fermi Launch• Launched on June 11, 2008• Science Operation on Aug 4,

2008• Orbit: 565 km, 26.5o (low BG)

Cape Canaveral Air Station @ Florida

5-yr mission(10-yr goal)

Fermi=LAT+GBM

15

Fermi_2011Mar.ppt

T. Mizuno et al.

Fermi-LAT Collaboration

France

Italy

Japan

Sweden

US

• Hiroshima Univ.• Tokyo Tech• ISAS/JAXA• Waseda Univ.• Tokyo Univ.• Nagoya Univ.• Aoyama Gakuin Univ.

~400 members

16

Fermi_2011Mar.ppt

T. Mizuno et al.

Large Area Telescope

• Pair-conversion type g-ray telescope

• ACD: plastic scintillatorsBG rejection

• Calorimeter: CsI scintillatorsEnergy measurement

• Tracker: Si-strip detectorsdirection measurement

20 MeV- 300 GeV

segmented tiles =>prevent self-veto

hodoscopic crystals =>shower profile

~200 um pitch =>high precision tracking

17

Fermi_2011Mar.ppt

T. Mizuno et al.

Large Area Telescope

• Pair-conversion type g-ray telescope

• Tracker: Si-strip detectorsdirection measurement

Key-element of LAT, developed by HPK and Hiroshima Univ.

Low-noise (~2.5nA/cm2)High-quality (dead strip: <0.01%)~106 channels in total

SSD

18

Fermi_2011Mar.ppt

T. Mizuno et al.

Performance of the LAT

• Large FOV (2.4 sr)• Large Aeff (>=8000 cm2 in 1-100 GeV)• Good PSF (0.6 deg@ 1GeV)

EGRET 3rd Catalog271 sources

Fermi-LAT 1st year catalog1451 sources

Atwood+09

sensitivity to point sources

19

Fermi_2011Mar.ppt

T. Mizuno et al.

Science, December 2009

• Fermi is recognized as one of the top science breakthroughs

Science Breakthroughs of 2009

Discovery of 16 new pulsars

20

Fermi_2011Mar.ppt

T. Mizuno et al.

GeV Gamma-ray Sky by LAT

• and provides us with high-quality data!– g-ray sources, diffuse g-rays and more

Fermi-LAT 1 year all-sky map

>100 publicationsas of Feb. 2011

21

Fermi_2011Mar.ppt

T. Mizuno et al.

Diffuse g-rays as probe of CRs and

ISM

22

Fermi_2011Mar.ppt

T. Mizuno et al.

GeV “Non” Excess

• Fermi does not confirm EGRET GeV excess

EGRETFermi-LAT

Abdo+09, PRL 103, 251101(CA: Johannesson, Porter, Strong)

local (<= 1 kpc from the sun)Diffuse Gamma-rays

E2 x F

lux

0.1 1 10 GeV

23

Fermi_2011Mar.ppt

T. Mizuno et al.

GeV “Non” Excess

• Fermi does not confirm EGRET GeV excess

EGRETFermi-LAT

Abdo+09(CA: Johannesson, Porter, Strong)

Instead, data is compatible with a standard model

E2 x F

lux

0.1 1 10 GeV

Gamma = CR x (ISM, ISRF)

G ~ 2.7

24

Fermi_2011Mar.ppt

T. Mizuno et al.

Diffuse gs in the Outer Galaxy

• Any unexpected in diffuse gs?– Yes. New information on CRs and ISM

The outer Galaxy (from outside)

sun

II quad III quad

L=90deg

L=180deg

L=270deg

25

Fermi_2011Mar.ppt

T. Mizuno et al.

Diffuse gs in the Outer Galaxy

• Any unexpected in diffuse gs?– Yes. New information on CRs and ISM

Abdo+10, ApJ 710, 133Ackermann+11, ApJ 726, 81 

The outer Galaxy (from the Fermi-LAT)

anticenterGC

26

Fermi_2011Mar.ppt

T. Mizuno et al.

ISM not visible by Standard Tracers

• Fermi revealed ISM gas not traced by radio surveys

Ackermann+11, ApJ 726, 81(CA: Grenier, Mizuno, Tibaldo)

Residual when fitted by N(HI)+CO

excess gammas= residual gas

27

Fermi_2011Mar.ppt

T. Mizuno et al.

ISM not visible by Standard Tracers

• Fermi revealed ISM gas not traced by radio surveys– confirming an earlier claim based on EGRET study

(Grenier+05)

Ackermann+11, ApJ 726, 81(CA: Grenier, Mizuno, Tibaldo)

Residual when fitted by N(HI)+CO Residual gas inferred by dust

28

Fermi_2011Mar.ppt

T. Mizuno et al.

More CRs than Expected

• Fermi detected more gs (more CRs) than a prediction based on SNR distribution and standard CR halo

Gam

ma-

ray

inte

nsity

Distance from GC (kpc)

Ackermann+11ApJ 726, 81(CA: Grenier, Mizuno, Tibaldo)

Sun

29

Fermi_2011Mar.ppt

T. Mizuno et al.

More CRs than Expected

• Fermi detected more gs (more CRs) than a prediction based on SNR distribution and standard CR halo– More CR sources or larger CR halo

Gam

ma-

ray

inte

nsity

Distance from GC (kpc)

Ackermann+11ApJ 726, 81(CA: Grenier, Mizuno, Tibaldo)

Sun

30

Fermi_2011Mar.ppt

T. Mizuno et al.

Remarks on Diffuse gs

• g-rays = CR x (ISM, ISRF)• “dark gas” is confirmed (more ISM gas)• More CRs than expected in outer Galaxy

• Improvement of diffuse g-ray modelBasis to search for anomalies

(in spectral and spatial distribution)

31

Fermi_2011Mar.ppt

T. Mizuno et al.

Behind Diffuse g-rays: Study of EGB and DM

search

32

Fermi_2011Mar.ppt

T. Mizuno et al.

Extragalactic Gamma-ray Background

• “Cosmic” Extragalactic Gamma-ray Background (EGB) has been known since 1970s

GeV background(EGRET)

CXB(resolved into AGNs)

Sreekumar+98

E2 x F

lux

keV MeV GeV

G ~ 2.1

33

Fermi_2011Mar.ppt

T. Mizuno et al.

Why is EGB Important?

• The EGB may encrypt the signature of the most powerful processes in astrophysics

Blazars contribute 20-100% of the EGB

Star forming galaxies

Particles accelerated in Intergalactic shocks

4% A

tom

s

Annihilation of Cosmological Dark Matter

Markevitch+05

Point sources

or diffuse

34

Fermi_2011Mar.ppt

T. Mizuno et al.

The Fermi EGB

• Fermi data + improved diffuse model– new EGB spectrum in 0.2-100 GeV

• carefully examine systematic uncertainty

Abdo+10, PRL 104, 101101 

+

+

=

LAT sky

gal. diffuse

point sources

Instrumenta

l BG

+”EGB”0.1 1 GeV 10 100

G ~ 2.4

Single PL, softer than EGRET result

(CA: Ackermann, Porter, Sellerholm)

35

Fermi_2011Mar.ppt

T. Mizuno et al.

Blazar Contribution

• Blazars account for a minimum of 16+-2%– Even if we extrapolate and integrate logN-logS

to zero, contribution is still <40%

logN-logS:Most of un-associated sources are likely to be blazars

0.1 1 GeV 10 100

flatter in F<=6x10-

8

Fermi EGB vs. source contribution

Abdo+10, ApJ 720, 435(CA: Ajello, Tramacere)

36

Fermi_2011Mar.ppt

T. Mizuno et al.

Blazar Contribution

• Blazars account for a maximum of 40% of EGB– g-ray “Fog” by Mysterious Dragons– star-forming galaxies, normal AGNs or truly diffuse?

0.1 1 GeV 10 100

Fermi EGB vs. source contribution

Abdo+10, ApJ 720, 435(CA: Ajello, Tramacere)

37

Fermi_2011Mar.ppt

T. Mizuno et al.

Limits on DM Annihilation

• Limits on DM by imposing the EGB is not violated

0.1 1 10 100 GeV

1.2 TeV m+m-

1.2x10-23 cm3/s200 GeV bb 5x10-25 cm3/s

180 GeV gg2.5x10-26 cm3/s

Abdo+10, JCAP 4, 14(CA: Conrad, Gustafsson, Sellerholm, Zaharijas)

38

Fermi_2011Mar.ppt

T. Mizuno et al.

Limits on DM Annihilation

• Limits on DM by imposing the EGB is not violated

10-22

10-25

<sv> (cm3/s)

100 1000GeVWIMP mass

w/o astrophysical sources

w/ astrophysical sources

Abdo+10, JCAP 4, 14(CA: Conrad, Gustafsson, Sellerholm, Zaharijas)

m+/m-

FermiPAMELA

39

Fermi_2011Mar.ppt

T. Mizuno et al.

Limits on DM Annihilation

• Limits on DM by imposing the EGB is not violated– already excluded some models, e.g., m+/m- channel favored

by PAMELA/Fermi e-/e+

10-22

10-25

<sv> (cm3/s)

100 1000GeVWIMP mass

w/o astrophysical sources

w/ astrophysical sources

Abdo+10, JCAP 4, 14(CA: Conrad, Gustafsson, Sellerholm, Zaharijas)

m+/m-

Fermi fit

40

Fermi_2011Mar.ppt

T. Mizuno et al.

Other Constraints on DM

Abdo+10, ApJ 712, 147(CA: Cohen-Tanugi, Farnier, Nuss, Profumo, Jeltema)

• Dwarf Spheroidal Galaxies are DM dominated– small BG (gas, star-forming activity)

dwarfs

m+/m- ULwith CMB IC

Diffusion coeff.

dwarfs

No detection

give constraints on some models, particularly m+/m- channel

41

Fermi_2011Mar.ppt

T. Mizuno et al.

Remarks on EGB and DM search

• “New” EGB spectrum in 0.2-100 GeV– Blazars account for <40% of EGB

• room for star-forming galaxies, normal AGNs or truly diffuse

• No evidence for DM annihilation– Constraints on models, in particular m+/m- channel– Astrophysical source contribution is important

42

Fermi_2011Mar.ppt

T. Mizuno et al.

Cosmic-Ray Electrons

43

Fermi_2011Mar.ppt

T. Mizuno et al.

PAMELA Positron Excess

• Convincing evidence of e+ ratio excess in >10 GeV– 2ndary e+ should be softer than primary e-

Adriani+08

Sources of “Primary” e+ are required

0.1 1 GeV 10 100

44

Fermi_2011Mar.ppt

T. Mizuno et al.

CRE by Fermi-LAT (2009)

• Fermi-LAT reported hard e- + e+ spectrum– Standard models with proper choice of params are able to

reproduce Fermi data alone, but not Fermi + PAMELA

Abod+09PRL 102, 181101(CA: Latronico, Moiseev)

Fermi Data (2009)

10 100 1000 GeV

6 months data4.5 M events

45

Fermi_2011Mar.ppt

T. Mizuno et al.

e-/e+ probe nearby sources

• CR e-/e+ loose energy via synchrotron and IC, hence probe nearby sources

We are here

e+ sources

Our Galaxy

KEK 井岡氏のトラぺより借用

46

Fermi_2011Mar.ppt

T. Mizuno et al.

CRE by Fermi-LAT (2010)

• CREs collected for 12 month (data is doubled)– Cross-check with events with long path in CAL (>=13X0)– LE extention using high latitude (low cutoff) data

• Noticeable deviation from a single PL

20 GeV

Fermi Data (2010)pre-Fermi model

Ackermann+10PRD 82, 092004(CA: Moiseev, Sgro)1 10 100 GeV 1000

47

Fermi_2011Mar.ppt

T. Mizuno et al.

CRE by Fermi-LAT (2010)

• Noticeable deviation from a single PL– Additional e-/e+ sources (astrophysical or extocis) can

provide a good fit to Fermi CRE and PAMELA e+/(e- + e+)

Example ofan additional component

Anisotropy of arrival direction may reveal sources or give constraints

1 10 100 GeV 1000

Ackermann+10PRD 82, 092004(CA: Moiseev, Sgro)

48

Fermi_2011Mar.ppt

T. Mizuno et al.

CRE AnisotropyNo-anisotropy map

Flight data sky map

Significance map

• Construct no anisotropy map from flight data– shuffling and direct integration

• Compare obtained map with data– search for anisotoropy

Ackermann+10PRD 82, 092003(CA: Mazziotta, Vasileiou)

No evidence of anisotropyin energies/angles investigated

Eth: 60-480 GeVAngular scale: 10-90 deg

49

Fermi_2011Mar.ppt

T. Mizuno et al.

Limit on Sources

• No evidence of anisotropies– Upper limit for the dipole anisotropy: 0.5-5%

• Limit already comparable to the value expected for a single nearby source dominating HE spectrum– will improve as more data are collected

Monogem

Vela

FermiHESS

Monogem

Vela

CRE spectrum at Earth Dipole Anisotropy

Fermi (3s UL)

Ackermann+10PRD 82, 092003(CA: Mazziotta, Vasileiou)

50

Fermi_2011Mar.ppt

T. Mizuno et al.

Summary

• Diffuse g-ray emission is a powerful probe for studying CRs and ISM

• Constraints on some DM models. Study of diffuse g-rays and g-ray object is important

• UL of CR anisotropy is close to what is expected for single nearby source

Thank you for your Attention

51

Fermi_2011Mar.ppt

T. Mizuno et al.

Summary

• 広がったガンマ線は , 宇宙線と星間ガスを研究する強力な手段 .

• DM探査で意味のある上限値が得られている . 広がった g線 + g線天体の理解は , DM探査にも資する .

• 電子陽電子源スペクトル +等方性も , 意味のある上限に近いところまできている .

Thank you for your Attention

52

Fermi_2011Mar.ppt

T. Mizuno et al.

Backup Slides

53

Fermi_2011Mar.ppt

T. Mizuno et al.

GeV “Non” Excess

Proton (pi0)

Electron (bremss) Electron (inverse Compton)

Isotropic (EGB+BG)

Abdo+090.1 1 10 GeV

E2 x F

lux

• instead, data is compatible with a model based on directly-measured CRs– solid basis to explore g-ray sky

54

Fermi_2011Mar.ppt

T. Mizuno et al.

More CRs than Expected

• Fermi detected more gs (= more CRs) than a prediction based on SNR distribution and standard CR halo.

LAT data vs.

standard model

Gam

ma-

ray

inte

nsity

Distance from GC (kpc)

Abdo+10, ApJ 710, 133(CA: Grenier, Tibaldo) Sun

55

Fermi_2011Mar.ppt

T. Mizuno et al.

e-/e+ probe nearby sources

• They loose energy via synchrotron and IC– dE/dt = -bE2

– T= 1/(bE) = 2.5x105 yr/(E/TeV)• hence are not able to reach far from the source

– R = (2DT)0.5 = 0.4-0.8 kpc @ 1TeV

• High Energy CR e-/e+ can probe nearby sources

D~(1-4)x1029 cm2/s@TeV

56

Fermi_2011Mar.ppt

T. Mizuno et al.

CRE by Fermi-LAT (2010)

• CREs collected for 12 month (data is doubled)– Cross-check with events with long path in CAL (>=13X0)– LE extention using high latitude (low cutoff) data

15.9 X0 avg.(1TeV shower peak @ 11 X0)

Consistent within their own systematics

@1TeVDE~14%DE~5%

100 GeV 1000