high energy gamma-ray emission from grbs masanori ohno(isas/jaxa) on behalf of fermi lat/gbm...

High EnergyGamma-ray Emission

from GRBs

Masanori Ohno(ISAS/JAXA)

on behalf of Fermi LAT/GBM collaborations

August 26, 2010 1ガンマ線バーストによるダークな宇宙の観測に向けたワークショップ

※MeV-GeVでの話です。

Outline• Introduction• Fermi Gamma-ray Space Telescope - Instruments - How does Fermi observe GRBs ?

• Science Result by Fermi - Extra component, delayed/Extended emission - Constraint on jet dynamics - other GRB topics (EBL, QG) - Upper limit on HE emission by Fermi data

• Summary

August 26, 2010 ガンマ線バーストによるダークな宇宙の観測に向けたワークショップ 2

Gamma-Ray BurstsGamma-Ray BurstsBright gamma-ray pulse in gamma-ray band is discovered in 1967

•GRBs originate from All-sky (~1GRBs/day)•Bimodal duration distribution:• non-thermal spectrum (Band function; synchrotron ?)•Short (<2s) and Long (>2s) GRB

BATSE (1991-)

BeppoSAX(1996-)• discovery of the X-ray afterglow

This leads a redshift measurement. cosmological origin for long GRBs(z=0.1-8)

relativistic jet is required (compactness problem)

HETE-2 (2002-) Swift (2004-)

•Leads many afterglow observations

•Association with SN and long GRBs

•Discovery of afterglow from short GRBs

Light curveLight curve

Duration distributionDuration distribution

coun

ts

20s

shortshort longlong

2s2s

most energetic explosion in the Universe (Eiso~1052 erg)


Gamma-Ray BurstsGamma-Ray BurstsBright gamma-ray pulse in gamma-ray band is discovered in 1967

•GRBs originate from All-sky (~1GRBs/day)•Bimodal duration distribution:• non-thermal spectrum (Band function; synchrotron ?)•Short (<2s) and Long (>2s) GRB

BATSE (1991-)

BeppoSAX(1996-)• discovery of the X-ray afterglow

This leads a redshift measurement. cosmological origin for long GRBs(z=0.1-8)

relativistic jet is required (compactness problem)

HETE-2 (2002-) Swift (2004-)

•Leads many afterglow observations

•Association with SN and long GRBs

•Discovery of afterglow from short GRBs

Light curveLight curve

Duration distributionDuration distribution

coun

ts

20s

shortshort longlong

2s2s

most energetic explosion in the Universe (Eiso~1052 erg)


α β

Epeak

10 100 1000Energy(keV)

Fv

vFv

Standard Model


Piran 2003

Prompt emission Afterglow

まだ謎は多い


emission mechanism, progenitor, short GRB…. etc

高エネルギーガンマ線 (>100MeV)の起源は特によく分かっていない。

8秒の周期性 ? SGR と関係 ?熱的放射 ? (Ryde & Pe’er 2009)

Iwakiri et al. in prep(4月京都会議より）

GRB 090709ABand function fit

BB + PL fit

vFv

vFv

100 1000 keV

HE emission from GRBs : Pre-Fermi EraHE emission from GRBs : Pre-Fermi Era

GRB940217v(Hurley et al. 94)

-18 to 14 sec

14 to 47 sec

47 to 80 sec

80-113 sec

113-211 sec

GRB941017 (Gonzaletz et al. 03)

GRB080514B AGILE

GeV photonsup to 90min after the trigger

Temporary distinct HE spectral component

Giuliani et al. 08Long-lived HE emission


HE emission from GRBs : Pre-Fermi EraHE emission from GRBs : Pre-Fermi Era


• 5 EGRET bursts with >50 MeV observations in 7 years– No evidence of cutoff in the summed spectrum

• Evidence for extended emission– >hour (afterglow?) GeV emission– Extra component at 100 s Composite spectrum of 5 EGRET Bursts

Dingus et al. 1997

No evidence of cutoff

What can we get from HE emission of GRBs?

Extra component of the prompt emission ?Extra component of the prompt emission ?Different emission mechanism: Synchrotron self Compton ? Hadronic origin ?Only GRB941017 shows the sign of extra component

What is the maximum energy of high energy photon?What is the maximum energy of high energy photon?

Constrain the bulk Lorentz factor of the relativistic jetNo evidence of the cut-off so far.

Delayed or long-lived high energy emission ?Delayed or long-lived high energy emission ?Suggests another emission mechanismTime delay of high energy photon Limit on the quantum gravity mass :MQG

A few GRBs show delayed high energy emission (GRB940217, GRB080714)

Need more sensitivity and larger FoV


Fermi Gamma-ray Space TelescopeFermi Gamma-ray Space Telescope

GGamma-ray amma-ray BBursturst MMonitor (onitor (GBMGBM)) 12 NaI detectors (8keV-1MeV) - onboard trigger , localization - spectroscopy 2 BGO detectors (150keV-40MeV) - spectroscopy (overlapping LAT

band)

LATLATSilicon-Strip detectors - Identification &direction measurement of γ-raysCsI calolimetor - Energy measurementACD (plastic scintillators) - background rejection


LAT Performance


100 ms27 usDeadtime per event

0.4 sr>2.2 srField of view

0.540.15Angular resolution (single photon, 10 GeV)

1500 cm29000 cm2Peak effective area

10%<10%Energy resolution (on axis, 100 MeV – 10 GeV)

20 MeV – 30 GeV20 MeV to >300 GeVEnergy range

EGRETLAT

• Major improvements in capabilities for GRB observation– Efficient observing mode (don’t look at Earth)– Wide FoV– Low deadtime (exploring dt’s down to µsec)

• Studies of short bursts possible– Large effective area– Good angular resolution– Increased energy coverage (to hundreds of GeV)

Many GRBs

More photons detected from each GRB Good GRB locations

Fermi GRB spectrum


GBM/NaI

GBM/BGO

LAT

10 107

Energy (keV)103 105

Alpha -1.02 +/- 0.02 Beta -2.21 +/- 0.03Epeak 1170 +/- 142 keV Amp. 0.0354 +/- 0.001 photons/s-cm2-keV REDUCED CHISQ = 0.963, PROB = 0.698

• Consistent with Band function from 10 keV to 10 GeV• No evidence for any other component • No evidence for any roll-off

(Abdo et al. 2009, Science 323, 1688)GRB 080916C

GBM/LAT on-board processing (10—15 s): GCN alert within 10—15 s from the trigger time through TDRSS (alert, location). GBM ground processing of prompt data (few minutes): Updated GBM position, preliminary light curve. LAT ground processing (a few hours after data downlink) Final location, spectrum (1st circular). Final location, high-energy flux and spectrum, afterglow search results (2nd circular).

Onboard Alert and Notices


Autonomous Repoint Request


Fermi S/C can be repointed for LAT observations of extended emission from bright GRBs

Track the target direction while above the horizon by at least the EarthAvoidance Angle (20. nominal), then slew at constant angle from the Earth limbuntil the the target rises on the other side.


GRB 090902B ARR

Burst Advocate


Fermi or other satellites detected GRB.

No

Notify analysis results to the team when the data is available.

1st GCN Circulars !!

Significant detection !!

Japanese members : ISAS/JAXA, Hiroshima U. and TITEC

LAT onboard alert ?

BA covers 24 hrs/day and 7day/week by an international collaboration.

Alert (E-mail)Alert (E-mail)

Yes

Within 30 min

~ a half of day (typically)


GRB 090902B: a successful observation sequence

Easily seen in 1 daycounts maps

• 11:05:15 UT = T0 – GBM GCN alert – ARR initiated– 15:05 UT – GBM human-in-the-loop localization

• LAT data monitoring and processing– 14:44 UT – GRB is seen in the telemetry– 18:24 UT – data ingest– 19:54 UT – GRB is seen in datamon plots

• 20:59:48 UT – FT1 file available [T0 + 10 hr]– ASP results ~20 min later, human-in-the-loop localization– Swift ToO request issued at ~21:30 UT, begins at 23:36 UT [T0 + 12.5 hr]

• 21:19:03 UT – 1st GBM circular (GCN 9866)• 22:48:18 UT – 1st LAT circular (GCN 9867)

– (RA,Dec=265.00, 27.33) with a 90% containment radius of 0.06 deg (statistical; 68% containment radius: 0.04 deg, systematic error <0.1 deg)

• 03:00:57 UT – Swift/XRT afterglow candidate (GCN 9868) [T0 + 16hr]– Estimated uncertainty of 4.2 arcseconds radius (90% confidence)– XRT position 3.2 arcmin from LAT position, inside error radius

• 04:57:44 UT – Swift/UVOT observations, no afterglow confirmation (GCN 9869)• 04:57:44 UT – enhanced Swift/XRT position (GCN 9871)• 07:36:42 UT – Fermi LAT and GBM refined analysis (GCN 9872)• 08:23:17 UT – Gemini-N absorption redshift (GCN 9873) z=1.822 (GMOS spectro) [T0 + 21.3 hr]• 09:14:50 UT – GROND localization 3.3 arcmin from LAT position

Fermi GRBs Fermi GRBs

August 26, 2010 18

Detections as of 090904

LAT FoV

GBM FoV

• The GBM detects ~250 GRBs/year (~400 total)– ~18% short– ~50% in the LAT FoV

• The LAT detects ~10 GRBs/year– 19 total as of today (recent detection : GRB 100724A)– ~10% of GBM GRBs observed

ガンマ線バーストによるダークな宇宙の観測に向けたワークショップ

• Only one case, LAT onboard notice has been issued (GRB090510) onboard algolythm has been updated, now we can expect 5 notices/year

LAT GRB detection rate

• ~9.3 GRBs/year with >10 photons >100 MeV• ~2.7 GRBs/year with >10 (100) photons >1 (0.1) GeV• Comparable to estimates based on Band spectrum fits to bright BATSE GRBs• Suggests that on average, GRBs don't have much excess (HE component) or deficit (cutoff) in the LAT energy range

w.r.t. the extrapolated Band spectrum from <2 MeV

Band et al., ApJ 701, 1673 (2009)

Extra PL component in short and long GRBs

Abdo, A. A. et al., ApJL 706, 138 (2009)Abdo, A. A. et al., ApJ submitted

GRB 090902B (long)GRB 090510 (short)

• First time a low-energy extension of the PL component has been seen

August 26, 2010 22

3 LAT GRBs shows extra PL component(090510, 090902B, 090926A)

First extra component by FermiAt > 5 sigma level


T0+4.6s to T0+9.6s

Limit on bulk Lorentz factorLimit on bulk Lorentz factor

Due to large luminosity and small emitting region, optical depth for the γ-γ -> e+e- pair production is too large to observe the non-thermal emission from GRB compactness problem.

Relativistic motion (Γ>>1) could avoid this compactness problem

Γmin can be derived using observed highest energy photon

August 26, 2010 24

Γmin~1000 for short and long GRBs

z

Γmin


090510E=31 GeV

090902BE=33 GeV

080916CE=3 GeV

GRB 090926A: the first HE spectral cutoff

Preliminary !

- Delay in HE onset: ~3 s- The extra component shows at >4 σ spectral break at ~1.4 GeV- First direct measurement of Γ ~ 720 (if cutoff due to γ-γ absorption)

August 26, 2010 25

8-14.3keV

14.3-260 keV

0.26-5 MeV

LAT all event

>100 MeV

>1GeV


Time-integrated photon spectrum(3.3-21.6s)

νFν(

erg

/cm

2/s

)

Energy (keV)10 102 103 104 105 106

Long-lived GeV emission ~ Swift and Fermi view of GRB 090510 ~

De Pasquale et al., ApJL 709, 146 (2010)

• Forward shock model can reproduce the spectrum from the optical up to GeV energies• Extensions needed to arrange the temporal properties

t1.380.07

Simultaneous fitof the SED at 5 different times

LAT emission until 200 sNo spectral evolution(photon index -2.1 ± 0.1)

August 26, 2010 27

GRB 090510 (short GRB)UVOT XRT Fermi/LAT


HE delayed onset in short and long GRBs

The first few GBM peaks are missing in the LAT but later peaks coincideDelay in HE onset: 0.1-0.2 s

Abdo et al. 2009, Science 323, 1688

The first LAT peak coincides with thesecond GBM peakDelay in HE onset: ~4-5 s

Abdo et al. 2009, Nature 462, 331

GRB 080916C (long)GRB 090510 (short)

HE delayed onset can be seen from almost all LAT GRBsAugust 26, 2010 28

8-260keV

0.26-5MeV

LAT all events

>100 MeV

>1GeV


Constraint on QG and EBL models


Constraints on the quantum gravity mass (MQG) by direct measurement of photon arrival time

MQG,1/Mplank > 1.19

Disfavors quantum gravity models which linearly alters the speed of light (n=1)

Most models are optically thin for33 GeV photon from GRB 090902B(z=1.822)

“baseline” and “fast evolution” modelsare rejected at 3.6 σ level

Abdo et al. 2009, Nature 462, 331GRB 090510 GRB 090902B

Abdo, A. A. et al., ApJL 706, 138 (2009)

31 GeV

GBM NaI

GBM BGO

LAT (>1MeV)

0.83 s

• Leptonic models (inverse-Compton or SSC) (Toma et al., 2009)

– Hard to produce a delayed onset longer than spike widths

– Hard to produce a low-energy (<50 keV) power-law excess

– Hard to account for the different photon index values of the Band spectrum at low energie (but photospheric models can) and of the HE component

– But, photospheric models could explain these properties (Toma et al. 2010)

• Hadronic models (pair cascades, proton synchrotron) (Asano et al., 2009)

– GRBs as possible sources of Ultra-High Energy Cosmic Rays

– Late onset: time to accelerate protons & develop cascades?

– Proton synchrotron radiation (requires large B-fields)

– Synchrotron emission from secondary e± pairs produced via photo-hadron interactions

• can naturally explain the power-law at low energies

• require substantially more energy than observed (GRB 090510: Etotal / Eiso ~ 100-1000)

– Hard to produce correlated variability at low- and high-energies (e.g. spikes of GRB 090926A) ?

• Early Afterglow (e+e- synchrotron from external shock) (Kumar et al, 2009)

– Can account for possible delayed (~9 s) onset of power-law component in GRB 090902B

– Short variability time scales in LAT data (e.g. GRB 090926A) argues against external shock

– Requires larger bulk Lorentz factor than measured for GRB 090926A

Models for HE delayed onset and extra-PL



Summary of LAT GRBs


GRBAngle from LAT

Duration(or class)

# of events> 100 MeV

# of events > 1 GeV

Delayed HE

onset

Long-lived HE

emission

Extra spectral comp.

Highest photon Energy

Redshift

080825C ~ 60° long ~ 10 0 ? ✔ X ~ 600 MeV

080916C 49° long 145 14 ✔ ✔ ? ~ 13.2 GeV ~ 4.35

081024B 21° short ~ 10 2 ✔ ✔ ? 3 GeV

081215A ~ 86° long — — — — -- —

090217 ~ 34° long ~ 10 0 X X X ~ 1 GeV

090323 ~ 55° long ~ 20 > 0 ? ✔ ? 3.57

090328 ~ 64° long ~ 20 > 0 ? ✔ ? 0.736

090510 ~ 14° short > 150 > 20 ✔ ✔ ✔ ~ 31 GeV 0.903

090626 ~ 15° long ~ 20 > 0 ? ✔ ?

090902B 51° long > 200 > 30 ✔ ✔ ✔ ~ 33 GeV 1.822

090926 ~ 52° long > 150 > 50 ✔ ✔ ✔ ~ 20 GeV 2.1062

091003A ~ 13° long ~ 20 > 0 ? ? ? 0.8969

091031 ~ 22° long ~ 20 > 0 ? ? ? ~ 1.2 GeV

100116A ~ 29° long ~ 10 3 ? ? ? ~ 2.2 GeV


Long vs Short GRBs (1)

August 26, 2010 32

PropertyProperty(HE: >0.1GeV)(HE: >0.1GeV) Short GRBs (2) Long GRBs (12)Delayed HE onset

1 out of 2 ✔(+ 1 likely but inconclusive)

3 out of 4 ✔ (+ 7 inconclusive cases)

Long-lived HE emission

2 out of 2 ✔ 7 out of 8 ✔(+ 3 inconclusive cases)

Redshift 1 out of 2 ✔(z = 0.903 for GRB090510)

6 out of 12 ✔(0.74, 0.90, 1.82, 2.11, 3.57, 4.35)

Bright 1 out of 2 ✔>100 (10) events >0.1 (1) GeV

3 out of 12 ✔>100 (10) events >0.1 (1) GeV

ΓΓmin min 1200 for GRB090510 900, 1000 (080916C, 090902B)

HE spectral component

1 out of 2 ✔(GRB090510)

2 out of 10 ✔(GRBs 090902B, 090926)

Similar high energy properties between short and long GRBsガンマ線バーストによるダークな宇宙の観測に向けたワークショップ

Long vs Short GRBs (2)

August 26, 2010 33

• Comparable LE and HE gamma-ray outputs for short GRBs• Long GRBs seem to emit ~5-20 times less at HE than at LE w.r.t.

short GRBs

short

Abdo, A. A. et al., ApJ 712, 558 (2010)

Preliminary !


short

short

Upper Limit on LAT GRB Flux


Fermi GRB statistics up to April 2010GBM Detections : 427 events - Within LAT FoV (<65 deg from LAT boresight) 213 events (~50 %) - LAT Detection 16 events (~7.5 %)

What can we say about remaining 197 events (92.5%)?

(1) LAT flux upper limit (2) Expected LAT flux by GBM extrapolation

•Intrinsic spectral break ?•EBL or γ-γ absorption ?

Possible Implication..

GBM Extrapolation


Bright GBM sample :• GBM events with > 70 BGO cts• But no LAT detections

Performed spectral fitting• 16 events in “Gold” sample• beta can be constrained

53 sample

Obtain expected LAT flux from GBM extrapolation

Compare LAT flux upper limit with expected LAT flux

Example: GRB 090620.400


Example: GRB 081207.680


SummarySummary

Fermi detected ~400 GRBs including 19 LAT GRBs in ~1.5 years => 250 GRBs/year for GBM and ~10 GRBs/year for LAT

April 19, 2010 38

Extra component of the prompt emission ?Extra component of the prompt emission ?

What is the maximum energy of HE emission ?What is the maximum energy of HE emission ?

HE emission is delayed and/or long-lived ?HE emission is delayed and/or long-lived ?

-Clear evidence of extra PL component from 3 LAT GRBs-Low-energy excess is also seen

-Constraint lower limit of bulk Lorentz factor: Γmin ~1000-GRB 090926A, first detection of HE spectral cutoff : Γ ~ 720

-Many LAT GRBs show delayed and long-lived high energy emission

Many leptonic or hadronic models are proposed for LAT high energy emission

No difference of high energy properties between short and long GRBs

Deciphering the Ancient Universe with Gamma-Ray Bursts

Constraint on QG and EBL models

high energy gamma-ray emission from grbs masanori ohno(isas/jaxa) on behalf of fermi lat/gbm...

Documents