grb x-ray afterglow observations with...

New Views of the Universe (Dec 2005) Abe Falcone, Penn State University

GRB X-ray Afterglow Observations with Swift-XRT

Abe Falcone(Pennsylvania State University)

and the Swift Team


XRT Collaborators

Penn State:David BurrowsJoanne HillJudith RacusinShiho KobayashiPeter MeszarosJohn NousekJamie KenneaDavid MorrisClaudio PaganiDirk GrupeAbe Falcone

OAB:Guido ChincariniGianpiero TagliaferriSergio CampanaAlberto MorettiPatrizia RomanoDaniele MalesaniStefano CovinoPaolo D’Avanzo

UL:Paul O’BrienAlan WellsJulian OsborneTony AbbeyAndy BeardmoreMike GoadKim PageDick Willingale

GSFC:Neil GehrelsLorella Angelini

UNLV:Bing Zhang

ASDC:Paolo GiommiMilvia Capalbi


Outline• XRT instrument & Burst follow-up Performance

(Ultra-brief)• Overall properties of Swift GRB afterglows

(a new multi-phase canonical picture emerges)

• The X-ray flares!

• Other equally interesting topics:

– Short GRB afterglows (see Fox et al. talk and Lamb plenary)

– High redshift afterglow (very briefly)

• Conclusions


•BAT First Light: 3 December 2004•XRT First Light: 11 December 2004

•First BAT Burst: 17 December 2004•First XRT Afterglow: 23 December 2004•UVOT First Light: 12 January 2005•Data public since 5 April 2005

The Swift Observatory

XRT First Light: Cas A

20 November 2004

Chandra First Light


BAT Bursts

• 92 GRBs detected/imaged since 12/17/05 (47 weeks as of 11/24/05) => 102/yr041217 050318 050412 050505 050607 050716 050801 050819

041219A,B,C 050319 050416A,B 050507 050701 050717 050802 050820A,B

041220 050326 050418 050509A,B 050712 050721 050803 050822

041223 050401 050421 050525 050713A,B 050724 050813 050824

041224 050406 050422 050528 050714B 050726 050814 050826

041226 050410 050502B 050603 050715 050730 050815 050827

041228

050117

050124

050126

050128

050202

050215A

050215B

050219A,B

050223

050306

050315

GRB Fluence

0.1

1

10

100

1000

12/1/

2004

1/1/20

052/1

/2005

3/1/20

054/1

/2005

5/1/20

056/1

/2005

7/1/20

058/1

/2005

9/1/20

0510

/1/20

0511

/1/20

05

Date

GR

B F

luen

ce (1

0^-7

)


XRT Detections of BAT GRBs

• Detected 71/77 = 92% with XRT (observed @ T < 200 ks)• Observed four during burst• 80% had prompt slews (< 350 s)• 53/56 = 95% of prompt XRT observations yielded detections

• 82% have fast decline or flare within first ~5 minutes

• 22 have redshift measurements:• Average redshift for LGRBs: 2.5

(compared with about 1 for Beppo-SAX bursts)• Highest redshift: 6.29

• SGRB mean redshift: 0.4

XRT Observation Start Times

02468

10

0 60 80 100

150

250

350

500

2000

1000

050

000

2000

0010

0000

0

Seconds since BAT Trigger

0

1

2

3

4

5

6

Freq

uenc

y

0 1 2 3 4 5 6 7

redshift


Swift Lightcurves – the Movie

Paul O’Brien / UL

BAT

XRT


The Overall Lightcurve

Zhang et al. 2005

Nousek et al. 2005


X-ray Flares: Some Background• Prior to Swift:

– Detection of X-ray flux increases from GRB011211 and GRB 011121 using Beppo-SAX (Piroet al. 2005)

– Conclusions were that flaring was onset of afterglow; not late-time internal shocks

– see also Zand et al. 2003, Galli & Piro 2005

• Source stacking studies of 100's of BATSE bursts; evidence for high energy emission well after nominal prompt emission phase (Connaughton2002)

• Could have been early evidence for flares; or it could have simply been the tail of the extrapolated nominal prompt emission; or it could have been the sum of both


How often do Swift GRBs have X-ray Flares?

~ 80% Swift XRT detections were prompt observations (< 300 s)

~1/3 of the prompt observations show flaring (>23 GRBs)


Myriad X-ray Flares050219A?050406050421050502B050607050712050713A050714B050716050724?050726050730050801050803050819050820A050822050904050908050916050922b051016b051117a

XRF 050406 GRB 050421 GRB 050502B

GRB 050712GRB 050607GRB 050713A

GRB 050714B GRB 050726GRB 050730


Flares: Motivation for Study

• Possible activity of internal GRB engine for long timescales can constrain energy sources and mechanisms

• In at least one case the fluence in an X-ray flare was comparable to initial prompt burst fluence

• Determine which flares are associated with internal engine and which are associated with afterglow/external shock

• Use timescales and other properties to constrain jet dynamics

• Investigate possible XRF-GRB connection


GRB 050502b: The Giant flare

Falcone, A. et al. 2005, ApJ, submittedBurrows, D. et al. 2005, Science

~500× increase!Rapid rise/decay!


GRB 050502b: The HUGE flare

• Flare fluence:(1.1 ± 0.05)×10-6 erg cm-2

• This is comparable to, or greater than, BAT prompt emission fluence

• Fast rise/decay (~t9)• Continuous power law

fits underlying afterglow, with no energy injection shift evident10

110

210

310

410

510

−2

10−1

100

101

102

103

Time Since BAT Trigger (sec)

XR

T C

ount

Rat

e (c

ount

s −

1 )



• VERY SHORT timescale variability near peak in hard band

• Hardness ratio increase and subsequent decay during flare

• During flare, Band function or cutoff power laws fit better than simple power law



• During flare, Band function or cutoff power laws fit better than simple power law

• Spectral index is harder during flare than that before and after, and a cutoff is indicated in the X-ray band


XRF 050406

• Weak burst, characterized as XRF, but with lightcurvecharacteristics very similar to those of GRBs with X-ray flares

• Lightcurve appears to level off– unresolved

rebrightenings?– energy injection into

afterglow shock?

• Flare δt/t ~ 0.3

102 103 104 105 106

Time since BAT trigger (s)

0.0001

0.0010

0.0100

0.1000

1.0000

10.0000

XR

T C

ou

nt

Rat

e (c

ou

nts

s-1)

XRF 050406A

Romano, P., et al. 2005, submitted


XRF 050406: Hardness

• Hardness ratio increase during flare, similar to GRB 050502b

0123

CR

(cp

s)

0123

CR

(cp

s)

0 100 200 300 400Time since BAT trigger (s)

0123

H/S

XRF 050406(S) 0.2-0.7 keV

(H) 0.7-10 keV

Romano, P., et al. 2005, submitted


Flares: Internal Vs. External• For many flares, there is no evidence for energy injection

into afterglow external shock at time of flare• Afterglow decay rate is consistent with afterglow having

already begun before flare, particularly for GRB 050502b• Small δt/t (<<1) for many flares makes it difficult for

external shock associated processes, particularly for GRB 050406

• For some flares, spectral index appears to change (harden) during flare, then it goes back to pre-flare value after the flare is over

• Bright flares can be fit better with Band function or cutoff power law, rather than simple power law


GRB 050820A – A simultaneous BAT/XRT flare

• Clear indication of energy injection into the forward shock

• A major, but only partially observed flare, with interesting spectrum

• z = 2.6


More X-ray Flare GRBs: 050716

courtesy of C. Hurkett

• Some flares occur during the steeper (region III) afterglow region

• Not ALL flares are dramatic sharp peaks


GRB 050904: z = 6.29Absorption redshift:

Subaru: 6.29 ± 0.01

Highest redshift GRB ever found.

Highest redshift X-ray emission ever detected.

2nd highest redshift ever measured (highest is 6.4).


GRB 050904: z = 6.29


GRB 050904: z = 6.29

Cusumano et al. 2005, Nature, submitted


GRB050724: Third Short GRB Afterglow

t90 = 1 s by BATSE definition. (But longsoft tail.)

30x brighter than GRB 050509B. (6E-7 ergs/cm2)

Slewed in 75 s. Very odd X-ray lightcurve.

Extremely steep decay!


Conclusions• Swift-XRT has successfully detected afterglows from >90% of BAT GRBs:

– New picture of early afterglow including a rapid decay that extrapolates back to BAT emission

– another decay phase that implies continued internal engine activity to late (>104 s) times

– relatively higher redshifts (mean ~2.7), including one z = 6.3 burst– First ever short burst afterglow detection

• X-ray flares during GRB afterglows are frequent and varied, with variety of peak times, ∆t/t, and relative magnitudes

– At least one flare shows evidence of variability on ~30 sec timescale, and ∆t/t<<1;==> very hard to model this with afterglow related mechanisms such as reverse shocks, IC emission, density variations,...

– Some flare spectra are fit better with Band function or cutoff power law, rather than simple power laws (note: this is important when calculating NH)

• Evidence suggests continued internal engine activity in at least some flares, probably late ejection episodes spawning internal shocks

• Several bursts exhibit multiple flares==> any model must involve multiple episodes


Flaring GRBs: The Movie


More X-ray Flare GRBs: 050713a

• X-ray flare captured during BAT prompt emission


More X-ray Flare GRBs: 050714b

figure courtesy K. Page

Possible continuation of prompt decay, with flare superimposed, followed by afterglow



• Incredibly fast flare rise: ~t14-t22!

• Early flaring (with fast ground-based response)

• fluence:~1.4×10-6 erg cm-2

C. Pagani et al. 2005, submitted



Courtesy of A. Moretti



courtesy D. Grupe

• Rising flux at start of XRT observations

• Followed by flares

•Still looking for a break at t > 106 s

• strong UVOT detection during early XRT flaring time region


Rosetta Stone(the burst that has it all!)

0

12

34

GRB 050315Vaughn et al. 2005

t -0.4t -0.7

ν -0.73±0.11

t -5.2

ν -1.9±0.9


GRB050509b: First Short GRB Afterglow

t90 = 0.04 s, Fluence = 2E-8 ergs/cm2

XRT counterpart in first 400 s, fades rapidly. 11 photons total.

Location in cluster at z=0.226, near early-type galaxy.

Possible NS-NS merger?

BAT: t-1.3

XRT: t-1.1

XRT error circle on VLT image. XRT position is 9.8”from a bright elliptical galaxy at z=0.226

Chandra

Gehrels et al. 2005, Nature

100x-1000x fainter than typical AG

grb x-ray afterglow observations with...

Documents