1. AGILE results on relativistic outflows above 100 MeV
Carlotta Pittori, ASI Space Science Data Center
on behalf of the AGILE Coll.
C. Pittori, INAF-OAR and SSDC HEPRO 6 in Moscow,

2. The AGILE Payload: the most compact instrument for HE astrophysics
only ~100 kg (~ 60  60 cm)
GRID gamma-ray imager (30 MeV - 30 GeV)
SuperAGILE hard X-ray imager ( keV)
MCAL Minicalorimeter ( MeV)
Italian Space Agency (ASI) Mission with INFN, INAF participation

3. Low Earth equatorial orbit: 550 Km and < 3º inclination angle
: AGILE is 10!
India April 23, 2007: Launch
Low Earth equatorial orbit: 550 Km and < 3º inclination angle

4. Malindi Ground Station
Satellite
Malindi Ground Station
Fucino TZP MOC
ASDC
AGILE Team
Guest Observers
Public data access
AGILE: “very fast” Ground Segment
~1 hr
~0.5 hr
Automatic
data processing
~0.5-1 hr
~(2-2.5) hr
Record for a gamma-ray mission! App AGILEScience for mobile dev
Now even faster (25 min latency) optimized for GW counterpart hunt!

5. ??? TERRA INCOGNITA ???

6. Recent discovery from HE astrophysics above 100 MeV
Gamma-ray intense transients on short timescales (< minutes, days) in different astrophysical systems:
Gamma-ray flares E>100 MeV in Microquasars (binary systems as Cyg X-3 and Cyg X-1)
Almost monochromatic gamma-ray flares from Crab Pulsar Wind Nebula
Gamma-ray flares up to TeV from FSRQ, with fast observed timescale variability of the order of minutes

7. High Energy Astrophysics: new lessons
Role of the magnetic field
Crab phenomenon (super-acceleration)
Plasma instabilities
Galactic compact objects (e. g. Cygnus X-3)
Quasars and relativistic jets, GRBs

8. The 3 results for which AGILE will be remembered:
Discovery of transient gamma-ray emission above 100 MeV from Cygnus X-3 microquasar in correlation with a repetitive pattern of multiwavelength (radio and X-ray) emission
Discovery of a new acceleration mechanism inducing intense and rapid flux variations in the Crab Nebula in the energy band above 100 millions of elettronvolt!
First direct evidence of cosmic ray acceleration in Supernovae remnants with the AGILE observations of the SNR W44 (2017 Matteucci Medal of the National Academy of Sciences to Marco Tavani) (topic not included in this talk)

9. Other results for which AGILE should be remembered:
The ‘‘Crazy Diamond’’ 3C 454.3, since August 2007 and other bright gamma-ray blazars
The remarkable short GRB (and the GW counterpart hunt)
Transient and subsequent discovery of the ‘‘hidden black hole’’ MCW 656 in a Be star binary (topic not included in this talk)
Surprises also from the Earth atmosphere: detection of Terrestrial Gamma-Ray Flashes (TGF) up to 100 MeV!! (topic not included in this talk)

10. Microquasars in the Cygnus region: Cygnus X-3 and Cygnus X-1
In this talk I will speak about AGILE observations of relativistic outflows at high energy from:
Microquasars in the Cygnus region: Cygnus X-3 and Cygnus X-1
Crab Nebula gamma-ray flares
Extreme flaring blazars: 3C and 3C 279
The short GRB as a template for Gravitational Waves counterpart hunt

11. Microquasars:
Galactic X-ray binary sources with radio emitting rel. jets
Open questions pre-AGILE and Fermi:
Can jet formation accelerate relativistic particles?
Can the jet emit γ-rays above 100 MeV?
The discovery of the γ-ray activity from Cygnus X-3 is the proof of extreme particle acceleration in microquasars!

12. Micro-quasars before the AGILE/Fermi era: The X-ray picture
Cygnus X-3
Cygnus X-1 γ?
S. Sabatini - Gamma-ray emission from Microquasars
(Zdziarski et al. 2002)
(Szostek et al., 2008)‏
Comptonization of soft thermal photons from disk by a hybrid population of electrons
(thermal + non-thermal) in the corona.
backward step
2 main X-ray spectral states:
Soft → thermal emission (BB)‏ from disk
+ Comptonization by cold thermal electrons (Soft Excess)
+ Comptonization by non-thermal high-energy electrons
(power-law tail)
Hard→ Comptonized emission by
hot quasi-thermal population of electrons
S. Sabatini

13. Slide adapted from G. Piano
The Cygnus region in γ-rays: AGILE Intensity Map (100 MeV-10 GeV) Pointing Mode: Nov – Oct. 2009, ~13 Ms net exposure time
SNR G
(Gamma Cygni)
AGILE GO pulsar (Halpern)
3 pulsars
3 TeV sources
2 microquasars
Slide adapted from G. Piano 13

14. The γ-ray detection of Cygnus X-3: brief story of a discovery
December 2, 2009:
The AGILE-GRID detects 4 γ-ray flares from Cygnus X-3
(“Extreme particle acceleration in the microquasar Cygnus X-3”, Tavani et al., Nature)
γ-ray flaring-fluxes greater than 1 order of magnitude with respect to the steady flux
coincident with prominent minima of the hard X-ray flux
a few days before major radio flares
December 11, 2009:
Fermi-LAT confirms AGILE detections
(“Modulated High-Energy Gamma-Ray Emission from the Microquasar Cygnus X-3”, Abdo et al., Science)
γ-ray detection of the orbital period (4.8 hours) → temporal signature of the microquasar
In 9 days a long-lasting mystery has been solved:
Cygnus X-3 is able to accelerate particles up to relativistic energies and to emit γ-rays above 100 MeV!

15. Major gamma-ray flares in special transitional states in preparation of radio flares! (Tavani et al, Nature 2009)
Gamma-ray flares tend to occur in the rare low-flux/pre-flare radio states.
For all gamma-ray flaring episodes, the radio and hard-X-ray fluxes are low or very low, while the soft X-ray flux is large
figure adapted from Szostek Zdziarski & McCollough (2008)

16. Cygnus X-1 Cygnus X-3 SPORADIC REPETITIVE PATTERN !!
Compact Object
4-15 M BH
1.4 M NS or 10 M BH
Companion
O9.7 Supergiant, L ~ 1039 erg/s
Wolf Rayet, L ~ erg/s
Companion wind
~ 10-6 M/yr, v ~ km/s
~ 10-5 M☉/yr, v ~ 1000 km/s
Period
5.6 days, orb. r. ~ 3.4 x 1012 cm
4.8 h, orb. r. ~ 3 x 1011 cm
Inclination Angle
30?
< 14
Sabatini et al. 2012
Tavani et al. Nature 2009
2 AGILE detected episodes in 2010
1 marginal FERMI detection in 2015
Hard state
Hard-to-soft transition
SPORADIC
REPETITIVE PATTERN !!
bright soft X-ray states (soft-to-hard state transitions)
state preceeding strong radio flares.

17. Cygnus X-3 lessons:
Direct evidence that extreme particle acceleration (above 100 MeV) and non-thermalized emission can occur in microquasars with a repetitive pattern (latest gamma-ray flares reported in March - April 2017, Fermi and AGILE ATels # 10109, 10138, 10179, 10243)
Emission must be produced not to far away from the central object (4,8 hours orbital modulation revealed by Fermi!)
Cyg X-3 is capable of accelerating particles by a very efficient mechanism leading to photon emission at energies thousands of times larger than the max energy previously detected (E ~ 300 keV)
Comptonization models (thermal and non-thermal) that reproduce the spectral states up to 300 keV must take into account the new data above 100 MeV

18. AGILE DISCOVERY OF THE CRAB NEBULA VARIABILITY IN -RAYS
MOST UNEXPECTED DISCOVERY FROM THE -RAY SKY:
AGILE DISCOVERY OF THE CRAB NEBULA VARIABILITY IN -RAYS
Tavani et al., Science, 331, 736 (2011)
Fermi confirmation:
Abdo et al., Science, 331, 739 (2011)

19. The Crab Nebula: a spectacular cosmic accelerator
THE STANDARD REFERENCE SOURCE IN ASTROPHYSICS
POWERFUL PULSAR (Neutron Star rotating 30 times a sec)
NEBULA SHOCKED BY THE PULSAR WIND
Crab Nebula: a remnant of a supernova that exploded in AD 1054 (Chinese astronomers). X-ray data from Chandra (light blue), visible light data from Hubble (dark blue and green) and infrared data from Spitzer (red), 31/1/2001
Marco Tavani, "AGILE Discovery of Gamma-Ray flares from the Crab Nebula" 19

20. Updated Crab Nebula spectrum from radio to TeV
Synchrotron burn-off
Synchrotron emission IC
Linear accelerator in ideal MHD framework
Diffusive acceleration
Eγ,max ≈ 25 MeV
Eγ,max = 9/4 mc2/α E/B ≈ 150 MeV E/B
Synchrotron burn-off (E/B<1)
(De Jager et al. 92, Arons 2012)
(Slide adapted from E. Striani, PhD Thesis)

21. The variable Crab Nebula!
FIRST PUBLIC ANNOUNCEMENT Sept. 22, 2010: AGILE issues the Astronomer’s Telegram n announcing a gamma-ray flare from the Crab Nebula
Science Express (6 January 2011)
Marco Tavani, "AGILE Discovery of Gamma-Ray flares from the Crab Nebula" 21

22. Crab Sept. 2010 flare: gamma-ray flare peak luminosity
L ≈ 5·1035 erg cm-2 s-1
kin. power fraction of PSR spindown Lsd,
ε ≈ (η-1/0.1) ≈ 0.01
timescales:
risetime: ≤ 1 day
decay: ~ 2-3 days
very efficient acceleration !
fast cooling,
B, Lorentz γ

23. Crucial constraints on shock particle acceleration theory !
e-/e+ shock acceleration by magnetic turbulence (diffusive vs. non-diffusive)
Crab Nebula shocks able to accelerate electrons/positrons at γ ~ 109 (PeV) !?
Variability never observed before within a day timescale

24. Big theoretical challenge:
Flare date
Duration
Peak γ-ray flux
Instruments
October 2007
~ 15 days
~ 6∙10-6 ph cm-2 s-1
AGILE
February 2009
~ 4∙10-6 ph cm-2 s-1
Fermi
September 2010
~ 4 days
~ 5∙10-6 ph cm-2 s-1
AGILE, Fermi
April 2011
~ 2 days
~ 30∙10-6 ph cm-2 s-1
Fermi, AGILE
Other -ray flaring states seen by Fermi and AGILE:
Mar and Oct 2013, Aug 2014, Oct Rate: ≈1/year
Big theoretical challenge:
the Crab Nebula is not a standard candle in gamma-rays! 24

25. Modelling of the April 2011 super-flare
AGILE
Fermi
No apparent relation between Optical, X-ray and Gamma-ray flaring emission
Mono-energetic (Maxwellian relativistic) distribution is favored
power-law
Maxwellian

26. Strong and impulsive flares (12-24 hr), ∼1/year
Striani et al., ApJ 2013
Strong and impulsive flares (12-24 hr), ∼1/year
Slower, less intense variability, and rather more frequent “waves” found in AGILE and Fermi data
Challenge to MHD models

27. South-East jet
2001
2010
Promising candidate for the gamma-ray flaring site?
Remarkable time variability and jet wiggling
High X-ray variability
Highly magnetized plasma
Kink instability in the jet could be responsible for the observed jet wiggling and trigger magnetic reconnection and particle acceleration.
3D rel. MHD simulations + resistivity: Mignone, Striani, et al

28. Crab lessons:
very exciting: the Crab Nebula is not a standard candle in gamma-rays
we “lost” the stability of an ideal reference source, but gained tremendous information about the fundamental process of particle acceleration
a big theoretical challenge, crucial feedback with laboratory plasma physics
the ultimate source of particle enhancements in the pulsar wind needs to be established: future surprises

29. Gamma-ray flaring blazars: 3C 454.3, 3C 279
Ratios between optical and -rays variation factors may be AAopt > 2 or more. Compton dominance varies
Gamma-ray only “orphan” flares. The Compton dominance attains values of 100 or more.
Very fast variability: -flux shows doubling times of few minutes
High energy spectrum can be unusually hard
(and PKS , PKS )

32. 3C 279 flares show different kinds of correlation opt - 
orphan
= 6
= 1
= 0
= 4
3C 279 flares show different kinds of correlation opt - 
The correlation is often absent.
Compton dominance rises to values > 100 in few hours.
Variations required in the external photon field
seen by the moving blob.
Mirrowing effect between plasmoids?
See Vittorini et al., ApJL 2017
(Fig. from Hayashida et al. 2015)

33. Donnarumma

34. AGILE and GW astrophysics
very fast reaction to external GW trigger
new processing pipelines
great potential for fast discovery of gamma-ray transients associated with NS-NS, NS-BH and BH-BH (if any) coalescences
AGILE GW-Team monitoring shifts (24/7) during the O2 GW LIGO-Virgo observing run.

35. First gravitational wave event: GW150914
T0 = 9:50:45 UT, 14 September, 2015
Announced by LIGO/Virgo on Feb. 11, 2016
We learned about the event on Feb. 11, (no MoU active yet): archival search
AGILE publication on the first GW event: Tavani et al., Ap.J. 825, L4 (2016) 35

36. AGILE-GRID in spinning: revolution including T0 of GW150914
Adapted from F. Verrecchia

37. AGILE field at T0 = 09:50:45 UT

38. AGILE just missed the T0 !

39. but only ~300 sec later…

40. AGILE exposure 330 sec (+/- 50 sec)

41. The “short” GRB 090510 lightcurve used as a possible HE template counterpart of the GW event
AGILE and the “short” GRB ”AGILE DETECTION OF DELAYED GAMMA-RAY EMISSION FROM THE SHORT GAMMA-RAY BURST GRB ” A. Giuliani et al., ApJL 708, 2010
z = 0.9

42. AGILE and the short GRB 090510 (61 degrees off-axis)
Delayed HE emission 1 2
GRID
counts
MCAL
counts
MCAL precursor < 1 MeV

43. AGILE and Fermi-LAT upper limits in the GRB 090510 light curve
(repositioned at z = 0.1, adapted from Fermi-LAT Collab., 2016)
AGILE-GRID: blu, Fermi/LAT: black, AGILE-GRID UL: red , AGILE UL extrapolation back to 1 s: green
Fermi-LAT UL
> 4500 s after T0
AGILE upper limit
sec after T0

44. 4π detector AGILE-MCAL and Fermi-GBM visibility regions at the GW150914 prompt time
Back: AGILE-GRID exposure map. Cross: Fermi-GBB candidate event location (about 0.4 s after T0)

45. AGILE-MCAL did not detect the transient reported by the Fermi GBM team
AGILE-GRID provided the most stringent constraint to any delayed emission above 50 MeV shortly after (~250 sec) the GW event
Great potential for AGILE observations of GW error boxes: prompt, minutes, hours, days

46. :
AGILE in the MoU promptly reacted to GW candidate events communicated by LIGO-Virgo
Several AGILE internal LVC-GCN Circulars were issued with reaction time of 2-3 hrs (including manual refined validation if data are available)
1 possible AGILE-MCAL gamma-ray transient candidate found as counterpart of the latest GW published event GW170104
Verrecchia et al. to appear in ApJL 2017, arXiv:

47. GW170104 – Jan. 4 2017 AGILE gamma-ray exp map
AGILE gamma-ray counts map
AGILE ASI Meeting, ASI HQ, Mar 9, 2017 47 47

49. MCAL candidate event E2 occurring 0
MCAL candidate event E2 occurring 0.46 sec before the final BH-BH coalescence time.
E2 very similar in timing and intensity to the famous 2009 precursor of GRB
E2 post-trial probability between 2.5 and 3 sigma (low signif., but similar to other claims)
being checked by other satellites
NEW MCAL pipeline developed for “sub- threshold events” (STE) btw 4 ÷ 5 sigma pre- trial

50. Let’s wait for the elephant!!
Thank you

52. EXTRA SLIDES

53. From S. Sabatini - Gamma-ray emission from Microquasars
Cygnus X-1
Cygnus X-3
Compact Object
4-15 M BH
1.4 M NS or 10 M BH
Companion
O9.7 Supergiant, L ~ 1039 erg/s
Wolf Rayet, L ~ erg/s
Companion wind
~ 10-6 M/yr, v ~ km/s
~ 10-5 M☉/yr, v ~ 1000 km/s
Period
5.6 days, orb. r. ~ 3.4 x 1012 cm
4.8 h, orb. r. ~ 3 x 1011 cm
Inclination Angle
30?
< 14
Cygnus X-1 (Cyg X-1) is one of the brightest X-ray sources in the sky. It is a binary system composed of a compact object and the O9.7 Iab supergiant star HDE , fill- ing 97 % of its Roche Lobe, with a mass ranging be- tween ∼ 15 and ∼ 30 M⊙ (see for example Gierlin ́ski et al. 1999; Caballero-Nieves et al. 2009). The measurement of the mass of the compact object, with a range between 4.8 M⊙ and 14.7 M⊙ by Herrero et al. (1995) and 8.7±0.8 M⊙ from Shaposhnikov & Titarchuk (2007), suggests identifi- cation with a black hole. The distance to the binary system is measured as 2.1 ± 0.1 kpc by Zi ́ol􏰀kowski (2005).
BECAUSE OF ITS BRIGHTNESS AND PERSISTENCY OF EMISSION, ITS OBSERVATIONAL AND THEORETICAL STUDIES HAVE BEEN OF GREAT IMPORTANCE FOR UNDERSTANDING THE PROCESS OF ACCRETION ONTO BLACK HOLES IN GENERAL
Cygnus X-3 is unique in orbital separation, luminosity of the companion star and inclination -> different behaviour can be expected in the two systems
From S. Sabatini - Gamma-ray emission from Microquasars

55. Vittorini, Tavani & Cavaliere: Mirrowing Effects in Flat Spectrum Radio Quasars

56. Non only gamma-rays from cosmic origin
but also: surprises from the EARTH atmosphere!!

57. TGF Cumulative spectrum
In 2011: 110 TGFs photons  E> 10 MeV  E> 20 MeV
Broken powerlaw model b = -2.7
Significant detection of 
>40 MeV!! Uneplained by standard RREA model: challenge for emission models
RREA cutoff powerlaw model
AGILE-MCAL crucial spectral contribution up to 100 MeV!!
Tavani et al., Phys. Rev. Letters 106, (2011) 57 57

58. AGILE GRB ON-BOARD SEARCH PROCEDURE
Super-A & MCAL
EVENTS
8 (SA) + 12 (MCAL)
DETECTOR
RATEMETERS (1 ms)
sub-ms
search
NORMAL BURST SEARCH
1 ms
16 ms
64 ms
256 ms
1.024 s
8.192 s
HARDWARE
HARDWARE
by dedicated
state-machine
SOFTWARE

59. AGILE two "lives": pointing and spinning
time period
Jul.07 – Oct.09
Nov today
attitude
fixed
variable
(rotation ~ 0.8º/sec)
sky coverage
1/5
~ %
1-day exposure ( 30 deg 100 MeV)
~ 2 x (cm2 sec)
( ) x (cm2 sec)