overview - fermi gamma-ray space...

Daniela Hadasch 5th Fermi Symposium

Overview

•  Introduction of LSI +61º 303 •  Present findings in the GeV energy regime

–  Using Fermi-LAT data

•  Putting results into multiwavelength context –  Comparing GeV data with radio, X-ray and optical data

•  Present possible interpretation of source behavior

2

Credit: Walt Feimer, NASA/Goddard Space Flight Center


Aragona et al. 2009

0.275

0.775

•  GeV source discovered in 1977 by Cos B –  at TeV energies by MAGIC (2006); follow-ups by MAGIC and VERITAS –  at GeV energies by Fermi (2009) –  Periodicity found in TeV (MAGIC) and GeV (Fermi/LAT)

•  Compact object + Be star –  Be star: ~10-12 Msun

–  B-type stars lose mass in equatorial, circumstellar disk –  Compact object: 1.4Msun (neutron star) to

4 Msun (black hole)

•  Orbital period –  (26.496 +/- 0.0028) days (Gregory et al. 2002)

•  Periodic radio outbursts (Gregory et al. 2002)

–  Superorbital period: (1667 +/- 8) days

•  What is happening at GeV energies?

Introduction: LSI +61º 303

3

Zabalza et al. 2011 A&A

periastron

apastron


Light folded in the orbital phase

4

0 0 2 0 4 0 6 0 8 1 1 2 1 4 1 6 1 8 2

]-1

s2 p

h cm

-6F

[10

0.5

1

1.5

Peria

stro

n

Apa

stro

n MJD 54683 - 54852

0 0 2 0 4 0 6 0 8 1 1 2 1 4 1 6 1 8 2

0.5

1

.5

Peria

stro

n

Apa

stro

n MJD 55529 - 55698

0 0 2 0 4 0 6 0 8 1 1 2 1 4 1 6 1 8 2

]-1

s2 p

h cm

-6F

[10

0.5

1

1.5 MJD 54852 - 55021

0 0 2 0 4 0 6 0 8 1 1 2 1 4 1 6 1 8 2

0.5

1

.5 MJD 55698 - 55867

0 0 2 0 0 6 0 8 1 1 2 1 1 6 1 8 2

]-1

s2 p

h cm

-6F

[10

0.5

1

1.5 MJD 55021 - 55191

0 0 2 0 0 6 0 8 1 1 2 1 1 6 1 8 2

0.5

1

.5 MJD 55867 - 56037

0 0 2 0 0 6 0 8 1 1 2 1 1 6 1 8 2

]-1

s2 p

h cm

-6F

[10

0.5

1

1.5 MJD 55191 - 55360

0 0 2 0 0 6 0 8 1 1 2 1 1 6 1 8 2

0.5

1

.5 MJD 56037 - 56206

Orbital Phase0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

]-1

s2 p

h cm

-6F

[10

0.5

1

1.5 MJD 55360 - 55529

Orbital Phase0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

0.5

1

.5 MJD 56206 - 56375

Each panel is ~6 months integration of Fermi data. Green and red background represents regions of periastron and apastron, respectively Trends for location of max and min Maximum near periastron, but with significant variability

Ackermann et al. 2013


10 half years shown for GeV data

5


Is there variability in the super orbit?

6

Time [MJD]55000 55500 56000

]-1

s-2

ph

cm-6

Flux

[10

0.6

0.7

0.8

0.9

1

1.1

Superorbital Phase0.8 1 1.2 1.4 1.6 1.8

•  Best determined superorbital period from radio campaign: (lasting 23 years): 1667±8 days à Probability that γ-ray flux evolution is a random result: < 1.1 × 10-12 à Source is variable along the superorbit in the GeV regime



Orbital phase bins in superorbit

7

]-1

s-2

ph

cm-6

F [1

0

0.5

1

1.5 orbital phase 0.0 - 0.1 orbital phase 0.5- 0.6

]-1

s-2

ph

cm-6

F [1

0

0.5

1

1.5 orbital phase 0.1 - 0.2 orbital phase 0.6 - 0.7

]-1

s-2

ph

cm-6

F [1

0

0.5

1


]-1

s-2

ph

cm-6

F [1

0

0.5

1


Superorbital Phase0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

]-1

s-2

ph

cm-6

F [1

0

0.5

1

1.5 orbital phase 0.4 - 0.5


orbital phase 0.9 - 1.0

Each panel shows the GeV flux at a fixed orbital position, along a period of 4.5 years Green and red background represent the region of periastron and apastron, respectively Black line: Fit sinusoidal with fixed superorbital period




8

]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1


]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1


]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1


]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1



]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1




•  From orbital phase 0.1 to

0.5, including the periastron region, there is no significant flux variation along the superorbit.

•  As soon as we depart from

periastron we start to see superorbital variability (see phase 0.5)

•  Conditions for GeV

generation must not significantly change



9

]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1

1.5 orbital phase 0.0 - 0.1 orbital phase 0.5- 0.6]

-1 s2

ph

cm-6

Flux

[10

0.5

1


]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1


]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1



]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1




•  From orbital phase 0.6 to 1.0, including the apastron region, there is significant flux variation in the superorbit.

•  The variation is maximal before and after apastron

•  Concurrently, a sine with a fixed period of 1667 days is at all orbital bins a better fit to the data than a constant

•  Close to apastron, the superorbit induces clear variations. GeV emission conditions change.



10

]-1

s-2

ph

cm-6

F [1

0

0.5

1


-1 s

-2 p

h cm

-6F

[10

0.5

1


]-1

s-2

ph

cm-6

F [1

0

0.5

1


]-1

s-2

ph

cm-6

F [1

0

0.5

1



]-1

s-2

ph

cm-6

F [1

0

0.5

1




•  From orbital phase 0.1 to

0.5, including the periastron region, there is no significant flux variation along the superorbit.

•  As soon as we depart from

periastron we start to see superorbital variability (see phase 0.5)

•  Conditions for GeV

generation must not significantly change




11

]-1

s-2

ph

cm-6

F [1

0

0.5

1


-1 s

-2 p

h cm

-6F

[10

0.5

1


]-1

s-2

ph

cm-6

F [1

0

0.5

1


]-1

s-2

ph

cm-6

F [1

0

0.5

1



]-1

s-2

ph

cm-6

F [1

0

0.5

1




•  From orbital phase 0.6 to 1.0, including the apastron region, there is significant flux variation in the superorbit.

•  The variation is maximal before and after apastron

•  Concurrently, a sine with a fixed period of 1667 days is at all orbital bins a better fit to the data than a constant

•  Close to apastron, the superorbit induces clear variations. GeV emission conditions change. Ackermann et al. 2013



12

]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1


]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1


]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1


]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1



]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1




Additional 1 year of data

confirms previous trend


RADIO DATA

13


GBI data, 8.3GHz, 1994-2000

14

Flux

(8.3

GHz

) [Jy

]

0

0.1

0.2

0.3

0.4

0.5 orbital phase 0.0 - 0.1 orbital phase 0.5- 0.6Fl

ux (8

.3G

Hz) [

Jy]

0

0.1

0.2

0.3

0.4


Flux

(8.3

GHz

) [Jy

]

0

0.1

0.2

0.3

0.4


Flux

(8.3

GHz

) [Jy

]

0

0.1

0.2

0.3

0.4



Flux

(8.3

GHz

) [Jy

]

0

0.1

0.2

0.3

0.4

0.5orbital phase 0.4 - 0.5



Same folding like in GeV à Modulation visible around apastron à Modulation starting earlier in orbital phase than in GeV?


GBI data, 2.25GHz, 1994-2000

15

Flux

(2.2

5GHz

) [Jy

]

0

0.1

0.2

0.3 orbital phase 0.0 - 0.1 orbital phase 0.5- 0.6Fl

ux (2

.25G

Hz) [

Jy]

0

0.1

0.2


Flux

(2.2

5GHz

) [Jy

]

0

0.1

0.2


Flux

(2.2

5GHz

) [Jy

]

0

0.1

0.2



Flux

(2.2

5GHz

) [Jy

]

0

0.1

0.2





GBI Data set: 2.25 GHz & 8.3 GHz

16

Norm

aliz

ed F

lux

0

1

2

3orbital phase 0.0 - 0.1 orbital phase 0.5- 0.6

Norm

aliz

ed F

lux

0

1

2

3orbital phase 0.1 - 0.2 orbital phase 0.6 - 0.7

Norm

aliz

ed F

lux

0

1

2


Norm

aliz

ed F

lux

0

1

2



Norm

aliz

ed F

lux

0

1

2

3orbital phase 0.4 - 0.5


orbital phase 0.9 - 1.0Blue: 2.25GHz Red: 8.3GHz

Average radio data in superorbital bins of 0.1 Both frequencies show almost same behavior Modulation visible around apastron, like in GeV


Radio (8.3GHz) - GeV

17

]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1


Jy]

-1Fl

ux (8

.3G

Hz)

[10

0

0.5

1

1.5

2orbital phase 0.5- 0.6

]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1


Jy]

-1Fl

ux (8

.3G

Hz)

[10

0

0.5

1

1.5


]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1


Jy]

-1Fl

ux (8

.3G

Hz)

[10

0

0.5

1

1.5


]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1


Jy]

-1Fl

ux (8

.3G

Hz)

[10

0

0.5

1

1.5



]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1



Jy]

-1Fl

ux (8

.3G

Hz)

[10

0

0.5

1

1.5


Blue: GeV Black: 8.3GHz

Pearson Corr. Coefficient: Correlation GeV – Radio on 3σ level around apastron


Radio (2.25GHz) - GeV

18

]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1


Jy]

-1Fl

ux (2

.25G

Hz)

[10

0

0.5

1

1.5


]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1


Jy]

-1Fl

ux (2

.25G

Hz)

[10

0

0.5

1

1.5


]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1


Jy]

-1Fl

ux (2

.25G

Hz)

[10

0

0.5

1

1.5


]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1


Jy]

-1Fl

ux (2

.25G

Hz)

[10

0

0.5

1

1.5



]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1



Jy]

-1Fl

ux (2

.25G

Hz)

[10

0

0.5

1

1.5


Blue: GeV Black: 2.25GHz


X-RAY

19


X-ray (RXTE): Orbital phase bins in superorbit

20

coun

ts/s

1

1.5

2

2.5

3 orbital phase 0.0 - 0.1 orbital phase 0.5- 0.6co

unts

/s

1

1.5

2

2.5

3 orbital phase 0.1 - 0.2 orbital phase 0.6 - 0.7

coun

ts/s

1

1.5

2

2.5


coun

ts/s

1

1.5

2

2.5



coun

ts/s

1

1.5

2

2.5

3 orbital phase 0.4 - 0.5



Modulation visible around apastron, like in GeV Black line: Sinusoidal fit with fixed superorbital period


Comparison X-ray - GeV

21

]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1


[cou

nts/

s](3

-30

keV)

Flux

1

2


]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1


[cou

nts/

s](3

-30

keV)

Flux

1

2


]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1


[cou

nts/

s](3

-30

keV)

Flux1

2


]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1


[cou

nts/

s](3

-30

keV)

Flux

1

2



]-1

s2 p

h cm

-6Fl

ux [1

0

0.5

1



[cou

nts/

s](3

-30

keV)

Flux1

2


Blue: GeV Red: X-ray

Pearson Corr. Coefficient: No significant correlation nor anticorrelation visible X-ray emission shifted wrt GeV?


OPTICAL DATA

22


Hα data over several years

23

Time [MJD]48000 49000 50000 51000 52000 53000 54000 55000 56000

]Å)[�

EW (H

0

5

10

15

20

25Paredes 1994Zamanov 1999Zamanov 2000, Fig. 1Zamanov 2013Liu 2000Liu 2005Grundstrom 2007McSwain 2010Jingzhi 2014

Equivalent width of Hα emission line also shows superorbital variability


Hα: Orbital phase bins in superorbit

24

No clear modulation at any part of the orbit visible Hα not a good tracer for characteristics of the disk?

]Å

) [�

EW(H

0

5

10

15

20orbital phase 0.0 - 0.1 orbital phase 0.5- 0.6

]Å

) [�

EW(H

0

5

10

15


]Å

) [�

EW(H

0

5

10

15


]Å

) [�

EW(H

0

5

10

15



]Å

) [�

EW(H

0

5

10

15





A possible interpretation for modulation

25

The superorbital variability in Be binary could be understood as a quasi-cyclical increase of the circumstellar disc size or

mass decretion rate The influence of the matter stripped off from the disk by the compact object’s passage can be larger and located farther out in periods of

higher mass loss

-2 -1.5 -1 -0.5 0 0.5 1 1.5 2-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

-2 -1.5 -1 -0.5 0 0.5 1 1.5 2-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

Periods of a relatively smaller disc

In periastron the influence of the cyclical increase of the disc is minor or nil, since the compact object is always affected by it. In apastron the influence of the cyclical increase of the disc is larger, but likely not maximal since the disc may not reach to overtake it.

Periods of a relatively larger disc

periastron

apastron


Summary - Outlook

•  Monitoring ongoing –  Behavior in GeV stays the same with additional one year of data

•  Multiwavelength behavior –  Radio: Also shows superorbital modulation distinguishing apastron from

periastron regions, like in GeV à Correlation at level of 3 sigma (Pearson corr. Coeff.)

–  X-ray: the same, plus no clear (anti)correlation visible. Emission shifted wrt GeV? –  Hα: no clear superorbital modulation visible when all data is considered together

à Hα not a good tracer? Too complex region mass-wise.

à Shown for first time that long term modulation visible at almost all frequencies à Radio, X-ray and GeV good tracer

•  Future plans –  Complete multi wavelength picture with TeV data –  Find tracer for disk behavior

26


THANK YOU

ARIGATŌ GOZAIMASU

有難う御座います

January 2014

overview - fermi gamma-ray space...

Documents