Embed Size (px)
Citation preview
ARTICLE IN PRESS
0168-9002/$ - se
doi:10.1016/j.ni
�CorrespondE-mail addr
Nuclear Instruments and Methods in Physics Research A 588 (2008) 52–62
www.elsevier.com/locate/nima
The AGILE space mission
M. Tavania,b,�, G. Barbiellinid, A. Argana, A. Bulgarellih, P. Caraveof, A. Chenc,f, V. Coccob,E. Costaa, G. De Parisa, E. Del Montea, G. Di Coccoh, I. Donnarummaa, M. Ferocia,
M. Fiorinif, T. Froyslandc,g, F. Fuschinoh, M. Gallii, F. Gianottih, A. Giulianif,Y. Evangelistaa, C. Labantih, I. Lapshova, F. Lazzarottoa, P. Liparij,k, F. Longod,e,
M. Marisaldih, M. Mastropietroa, F. Mauril, S. Mereghettif, E. Morellih, A. Morsellig,L. Pacciania, A. Pellizzonif, F. Perottif, P. Picozzab,g, C. Pontonid, G. Porrovecchioa,
M. Prestd, G. Pucellaa, M. Rapisardam, E. Rossih, A. Rubinia, P. Soffittaa,M. Trifoglioh, A. Troisf, E. Vallazzad, S. Vercellonef, A. Zambrac,f, D. Zanelloj,k,
P. Giommin, A. Antonellin, C. Pittorin
aINAF-IASF Roma, via del Fosso del Cavaliere 100, I-00133 Roma, ItalybDip. Fisica, Universita Tor Vergata, via della Ricerca Scientifica 1, I-00133 Roma, Italy
cCIFS, villa Gualino - v.le Settimio Severo 63, I-10133 Torino, ItalydINFN Trieste, Padriciano 99, I-34012 Trieste, Italy
eDip. Fisica, Universita di Trieste, via A. Valerio 2, I-34127 Trieste, ItalyfINAF-IASF Milano, via E. Bassini 15, I-20133 Milano, Italy
gINFN Roma 2, via della Ricerca Scientifica 1, I-00133 Roma, ItalyhINAF-IASF Bologna, via Gobetti 101, I-40129 Bologna, ItalyiENEA Bologna, via don Fiammelli 2, I-40128 Bologna, Italy
jINFN Roma 1, p.le Aldo Moro 2, I-00185 Roma, ItalykDip. Fisica, Universita La Sapienza, p.le Aldo Moro 2, I-00185 Roma, Italy
lINFN Pavia, via Bassi 6, I-27100 Pavia, ItalymENEA Frascati, via Enrico Fermi 45, I-00044 Frascati (RM), ItalynASI Science Data Center, ESRIN, I-00044 Frascati (RM), Italy
Available online 17 January 2008
Abstract
AGILE is an Italian Space Agency mission dedicated to the exploration of the gamma-ray Universe. The AGILE, very innovative
instrument, combines for the first time a gamma-ray imager (sensitive in the range 30MeV–50GeV) and a hard X-ray imager (sensitive in
the range 18–60 keV). An optimal angular resolution and very large fields of view are obtained by the use of state-of-the-art Silicon
detectors integrated in a very compact instrument. AGILE was successfully launched on April 23, 2007 from the Indian base of
Sriharikota and was inserted in an optimal low-particle background equatorial orbit. AGILE will provide crucial data for the study of
Active Galactic Nuclei, Gamma-Ray Bursts, unidentified gamma-ray sources, galactic compact objects, supernova remnants, TeV
sources, and fundamental physics by microsecond timing. The AGILE Cycle-1 pointing program started on 2007 December 1, and is
open to the international community through a Guest Observer Program.
r 2008 Elsevier B.V. All rights reserved.
PACS: 95.55.�n; 95.85.Pw; 95.85.Nv
Keywords: Astronomical and space-research instrumentation; Astronomical observations; X-ray and gamma-ray
e front matter r 2008 Elsevier B.V. All rights reserved.
ma.2008.01.023
ing author at: INAF-IASF Roma, via del Fosso del Cavaliere 100, I-00133 Roma, Italy.
ess: [email protected] (M. Tavani).
ARTICLE IN PRESSM. Tavani et al. / Nuclear Instruments and Methods in Physics Research A 588 (2008) 52–62 53
1. Introduction
The space program AGILE (Astro-rivelatore Gamma a
Immagini LEggero) is a high-energy astrophysics missionsupported by the Italian Space Agency (ASI) with scientificand programmatic participation by INAF, INFN andseveral Italian universities [1]. The industrial team includesCarlo Gavazzi Space, Thales-Alenia-Space-Laben, Oerli-kon-Contraves, and Telespazio.
The main scientific goal of the AGILE program is toprovide a powerful and cost-effective mission with excellentimaging capability simultaneously in the 30MeV–50GeVand 18–60 keV energy ranges with a very large field of view(FOV) [2–4].
AGILE was successfully launched by the Indian PSLV-C8rocket from the Sriharikota base on April 23, 2007 (Fig. 1).The launch and orbital insertion were nominal, and a quasi-equatorial orbit was achieved with the smallest inclination(2.51) ever achieved by a high-energy space mission. Thesatellite commissioning phase was carried out during theperiod May–June, 2007. The in-orbit calibration, based onlong pointings at the Vela and Crab pulsars was carried outduring the period July–October, 2007. The nominal observa-tion phase (AGILE Cycle-1) started on December 1, 2007.
Fig. 1. The launch of the AGILE satellite by the Indian PSLV-C8 rocket
from the Sriharikota base on April 23, 2007.
The AGILE instrument design is innovative andbased on the state-of-the-art technology of solid stateSilicon detectors and associated electronics developed inItalian laboratories [5–9]. The instrument is very compact(see Fig. 2) and light (�120 kg) and is aimed at detectingnew transients and monitoring gamma-ray sources within avery large FOV (�1
5of the whole sky). The total satellite
mass is equal to 350 kg (see Fig. 3).AGILE is expected to substantially advance our knowl-
edge in several research areas including the study of ActiveGalactic Nuclei and massive black holes, Gamma-RayBursts (GRBs), the unidentified gamma-ray sources,galactic transient and steady compact objects, isolatedand binary pulsars, pulsar wind nebulae (PWNae), super-nova remnants, TeV sources, and the Galactic Center.Furthermore, the fast AGILE electronic readout and dataprocessing (resulting in detectors’ deadtimes smaller than�200ms) allow for the first time a systematic search forsub-millisecond gamma-ray/hard X-ray transients that areof interest for both galactic objects (searching outburstdurations comparable with the dynamical timescale of�1M� compact objects) and quantum gravity studies [10].The AGILE Science Program is focused on a prompt
response to gamma-ray transients and alert for follow-upmultiwavelength observations. AGILE will provide crucialinformation complementary to several space missions(Chandra, INTEGRAL, XMM-Newton, SWIFT, Suzaku)and it will support ground-based investigations in theradio, optical, and TeV bands. Part of the AGILE ScienceProgram will be open for Guest Investigations on acompetitive basis. Quicklook data analysis and fastcommunication of new transients will be implemented asan essential part of the AGILE Science Program.
Fig. 2. The AGILE scientific instrument showing the hard X-ray imager,
the gamma-ray Tracker, and Calorimeter. The Anticoincidence system is
partially displayed, and no lateral electronic boards and harness are shown
for simplicity. The AGILE instrument ‘‘core’’ is approximately a cube of
about 60 cm size and of weight equal to 100 kg.
ARTICLE IN PRESS
Fig. 3. The integrated AGILE satellite in its final configuration before
being covered by the thermal blanket. The total satellite mass is equal to
350 kg.
M. Tavani et al. / Nuclear Instruments and Methods in Physics Research A 588 (2008) 52–6254
2. The scientific instrument
The AGILE scientific payload is made of three detectorscombined into one integrated instrument with broad-banddetection and imaging capabilities. The Anticoincidenceand Data Handling (DH) systems complete the instrument.
The gamma-ray imaging detector (GRID) is sensitive inthe energy range �30MeV–50GeV, and consists of aSilicon–Tungsten Tracker, a Cesium Iodide Calorimeter,and the Anticoincidence system.1 It is characterized by avery fine spatial resolution (obtained by a special arrange-ment of Silicon microstrip detectors and analog signalstorage and processing) and by the smallest ever obtaineddeadtime for gamma-ray detection (t200ms). The GRIDis designed to achieve an optimal angular resolution(source location accuracy �150 for intense sources), anunprecedentedly large FOV (�2:5 sr), and a sensitivitycomparable to that of EGRET for sources within10–201 off-axis (and substantially better for larger off-axisangles).
1In contrast with previous generation instruments (COS-B, EGRET),
AGILE does not require gas operations and/or refilling, and does not
require high voltages.
The hard X-ray imager (Super-AGILE) is a uniquefeature of the AGILE instrument. This imager is placed ontop of the gamma-ray detector and is sensitive in the18–60 keV band. It has an optimal angular resolution(6 arcmin) and a good sensitivity over a �1 sr FOV(�10–15mCrab on-axis for a 1-day integration). The maincharacteristic of AGILE will be then the possibility ofsimultaneous gamma-ray and hard X-ray source detectionwith arcminute positioning and on-board GRB/transientsource alert capability.
A Mini-Calorimeter operating in the ‘‘burst mode’’ is the
third AGILE detector. It is part of the GRID, but also iscapable of independently detecting GRBs and othertransients in the 350 keV–100MeV energy range withoptimal timing capabilities.Fig. 3 shows the integrated AGILE satellite and Fig. 2 a
schematic representation of the instrument. We brieflydescribe here the main detecting units of the AGILEinstrument; more detailed information will be presentedelsewhere [11].
�
Fig
lab
ins
The Silicon-Tracker (ST) providing the gamma-rayimager is based on photon conversion into electron–po-sitron pairs. It consists of a total of 12 trays with arepetition pattern of 1.9 cm (Fig. 4). The first 10 traysare capable of converting gamma-rays by a Tungstenlayer. Tracking of charged particles is ensured by high-resolution Silicon microstrip detectors that are config-ured to provide the two orthogonal coordinates for eachelement (point) along the track. The fundamentalSilicon detector unit is a tile of area 9:5� 9:5 cm2,microstrip pitch equal to 121mm, and thickness 410mm.The AGILE ST readout system is capable of detectingand storing the energy deposited by the penetratingparticles for half of the Silicon microstrips. This impliesan alternating readout system characterized by ‘‘read-out’’ and ‘‘floating’’ strips. The analog signal producedin the readout strips is read and stored for further
. 4. The assembled AGILE Silicon Tracker developed by the INFN
oratories of Trieste before being integrated with the rest of the
trument (June 2005).
ARTICLE IN PRESS
Fig
by
clu
Fig. 6. The first gamma-ray event detected by AGILE in space (May 10,
2007).
M. Tavani et al. / Nuclear Instruments and Methods in Physics Research A 588 (2008) 52–62 55
processing. The fundamental element is a Silicon tilewith a total of 384 readout channels (readout pitch equalto 242mm) and three TAA1 chips required to processindependently the analog signal from the readout strips.Each Si-Tracker layer is made of 4� 4 Si-tiles, for atotal geometric area of 38� 38 cm2. The first 10 traysare equipped with a Tungsten layer of 245mm (0:07X 0)positioned in the lower part of the tray. The Silicondetectors providing the two orthogonal coordinates arepositioned at the very top and bottom of these trays.For each tray there are 2� 1536 readout microstrips.Since the ST trigger requires at least three planes to beactivated, two more trays are inserted at the bottom ofthe Tracker without the Tungsten layers. The totalreadout channel number for the GRID Tracker is then36,864. Both digital and analog information (chargedeposition in Si-microstrip) is read by TAA1 chips. Thedistance between mid-planes equals 1.9 cm (optimizedby Monte Carlo simulations). The ST has an on-axis
total radiation length near 0:8X 0. Special trigger logicalgorithms implemented on-board (Level-1 and Level-2)lead to a substantial particle/albedo-photon backgroundsubtraction and a preliminary on-board reconstructionof the photon incidence angle. Both digital andanalog information are crucial for this task. Fig. 5shows a typical gamma-ray event detected by the GRIDduring the pre-launch satellite tests. Fig. 6 shows thefirst gamma-ray event detected by the GRID in orbit.
. 5. A typical gamma-ray event produced by cosmic-rays and detected
the AGILE Tracker with its characteristic pattern of released energy in
sters of hit readout Silicon microstrips.
Fig. 7. The Super-AGILE detector during metrology measurements
(March 2005).
The positional resolution obtained by the ST isexcellent, being below 40 mm for a large range of particleincidence angles [12].
� Super-AGILE (SA), the ultra-compact and light hard-X-ray imager of AGILE [13] is a coded-mask systemmade of a Silicon detector plane and a thin Tungstenmask positioned 14 cm above it (Fig. 7). The detectorplane is organized in four independent square Silicondetectors (19� 19 cm2 each) plus dedicated front-endelectronics based on the XAA1.2 chips (suitable in the SAenergy range). The total number of SA readout channelsis 6144. The detection cabability of SA includes: (1)photon-by-photon transmission and imaging of sources
ARTICLE IN PRESSM. Tavani et al. / Nuclear Instruments and Methods in Physics Research A 588 (2008) 52–6256
in the energy range 18–60 keV, with a large FOV (�1 sr);(2) a good angular resolution (6 arcmin); (3) a goodsensitivity (�10mCrab between 18 and 60keV for 50ksintegration, and t1Crab for a few seconds integration).The hard-X-ray imager is aimed at the simultaneousX-ray and gamma-ray detection of high-energy sourceswith excellent timing capabilities (a few microseconddeadtime for individual detectors). The AGILE satellite isequipped with an ORBCOMM transponder capable oftrasmitting GRB coordinates to the ground within2–3min.
� The Mini-Calorimeter (MCAL) is made of 30 Thalliumactivated Cesium Iodide (CsI(Tl)) bars arranged in twoplanes, for a total (on-axis) radiation length of 1:5X 0.The signal from each CsI bar is collected by twophotodiodes placed at both ends. The MCAL aims are:(i) obtaining additional information on the energydeposited in the CsI bars by particles produced in theSilicon Tracker (and therefore contributing to thedetermination of the total photon energy); (ii) detectingGRBs and other impulsive events with spectraland intensity information in the energy band�0:35–100MeV. An independent burst search algorithmis implemented on board with a wide dynamic range forthe MCAL independent GRB detection.
� The Anticoincidence (AC) system is aimed at bothcharged particle background rejection and preliminarydirection reconstruction for triggered photon events.The AC system completely surrounds all AGILEdetectors (Super-AGILE, Si-Tracker and MCAL). Eachlateral face is segmented in three plastic scintillatorlayers (0.6 cm thick) connected with photomultipliersplaced at the bottom. A single plastic scintillator layer(0.5 cm thick) constitutes the top-AC whose signal isread by four light photomultipliers placed externally tothe AC system and supported by the four corners of thestructure frame. The segmentation of the AC system andthe ST trigger logic contribute to produce the very largeFOV of the AGILE-GRID.
The DH and power supply systems complete theinstrument. The DH is optimized for fast on-boardprocessing of the GRID, Mini-Calorimeter and Super-AGILE data. Given the relatively large number of readablechannels in the ST and Super-AGILE (�40; 000), theinstrument requires a very efficient on-board data proces-sing system. The GRID trigger logic for the acquisition ofgamma-ray photon data and background rejection isstructured in two main levels: Level-1 and Level-2 triggerstages. The Level-1 trigger is fast (t5ms) and requires asignal in at least three out of four contiguous trackerplanes, and a proper combination of fired TAA1 chipnumber signals and AC signals. An intermediate Level-1.5stage is also envisioned (lasting �20ms), with the acquisi-tion of the event topology based on the identification offired TAA1 chips. Both Level-1 and Level-1.5 have ahardware-oriented veto logic providing a first cut of
background events. Level-2 data processing includes aGRID readout and pre-processing, ‘‘cluster data acquisi-tion’’ (analog and digital information). The Level-2processing is asynchronous (estimated duration � a fewms) with the actual GRID event processing. The GRIDdeadtime turns out to be t200ms and is dominated by theTracker readout.The charged particle and albedo-photon background
passing the Level-1þ 1:5 trigger level of processing issimulated to be t100 events=s for the nominal equatorialorbit of AGILE. The on-board Level-2 processing has thetask of reducing this background by a factor between 3and 5. Off-line processing of the GRID data with bothdigital and analog information is being developed with thegoal to reduce the particle and albedo-photon backgroundrate above 100MeV to �0:01 events=s.In order to maximize the GRID FOV and detection
efficiency for large-angle incident gamma-rays (and mini-mize the effects of particle backsplash from the MCAL andof ‘‘Earth albedo’’ background photons), the data acquisi-tion logic uses proper combinations of top and lateral ACsignals and a coarse on-line direction reconstruction inthe ST. For events depositing more than 200MeV in theMCAL, the AC veto may be disabled to allow theacquisition of gamma-ray photon events with energieslarger than 1GeV.A special set of memory buffers and burst search
algorithms are implemented to maximize data acquisitionfor transient gamma-ray events (e.g., GRBs) in the ST,Super-AGILE and Mini-Calorimeter, respectively. TheSuper-AGILE event acquisition envisions a first ‘‘filtering’’based on AC-veto signals, and pulse-height discriminationin the dedicated front end electronics (based on XA1chips). The events are then buffered and transmitted tothe CPU for burst searching and final data formatting.The four Si-detectors of Super-AGILE are organized in16 independent readout units, resulting in a �5ms globaldeadtime.In order to maximize the detecting area and minimize the
instrument weight, the GRID and Super-AGILE front-end-electronics is partly accommodated in special boardsplaced externally on the Tracker lateral faces. Electronicboxes, P/L memory (and buffer) units are positioned at thebottom of the instrument within the spacecraft body.
3. Science with AGILE
This section summarizes the main features of the AGILEscientific capability. The AGILE instrument has beendesigned and developed to obtain:
�
excellent imaging capability in the energy range100MeV–50GeV, improving the EGRET angularresolution by a factor of 2, see Fig. 9;
� a very large FOV for both the gamma-ray imager (2.5 sr)and the hard X-ray imager (1 sr), (FOV larger by afactor of �6 than that of EGRET);
ARTICLE IN PRESSM. Tavani et al. / Nuclear Instruments and Methods in Physics Research A 588 (2008) 52–62 57
�
-1-1
Fig
a f
Cra
excellent timing capability, with overall photon absolutetime tagging of uncertainty below 2ms and very smalldeadtimes (t200ms for the GRID, �5ms for the sum ofthe SA readout units, and �20ms for each of theindividual CsI bars);
� a good sensitivity for point sources, comparable to that ofEGRET in the gamma-ray range for sources within201 off-axis (except a central region of smaller effectivearea), and very flat up to 50–601 off-axis (see Fig. 8).Depending on exposure and the diffuse background, theflux sensitivity threshold can reach values of ð10220Þ �10�8 ph=cm2=s above 100MeV. The hard X-ray imagersensitivity is between 10 and 20mCrab at 20 keV for a1-day integration over a FOV near 1 sr;
�Fig. 9. Simulated positioning of a gamma-ray source in the galactic plane
good sensitivity to photons in the energy range
�30–100MeV, with an effective area above 200 cm2 at30MeV;
obtained by the AGILE Likelihood and diffuse gamma-ray background
� processing. The (more compact) red contours show the 50%, 68%, 95%,99% confidence levels obtained for the AGILE-GRID; the black contours
a very rapid response to gamma-ray transients and GRBs,obtained by a special quicklook analysis program andcoordinated ground-based and space observations;
show the confidence levels obtained with the EGRET likelihood analysis.
� accurate localization (�2–3 arcmins) of GRBs and othertransient events obtained by the GRID-SA combination(for typical hard X-ray transient fluxes above �1Crab);the expected GRB detection rate is �1–2 per month;
� long-timescale continuous monitoring (�2–3 weeks) ofgamma-ray and hard X-ray sources;
� satellite repointing after special alerts (�1 day) in orderto position the source within the Super-AGILE FOV toobtain hard X-ray data for gamma-ray transientsdetected by the GRID in the external part of its FOVor by other high-energy missions (e.g., GLAST).
The combination of simultaneous hard X-ray andgamma-ray data will provide a formidable combinationfor the study of high-energy sources. We briefly address
10-2 100 102 104
Energy [MeV]
30 degrees off-axisPart. bkg.: 0.01-0.1 Hz20% occultation
102
100
10-2
10-4
10-6
10-8
Sen
sitiv
ity (>
E) [
ph.c
ms
]
2 days
2 days2 weeks1 year
Super-AGILE AGILE-GRID
Super-AGILE, AGILE-GRID |b|<10 deg Sens.
. 8. Simulated integrated flux sensitivity of AGILE (GRID and SA) as
unction of energy for a 30� off-axis source in the galactic plane. The
b spectrum is shown by the dotted line.
here some of the relevant features of the expected scientificperformance (Fig. 9).
3.1. Angular resolution
The GRID configuration will achieve a PSF with 68%containment radius better within 122� at E4300MeVallowing a gamma-ray source positioning with error boxradius near 100–200 depending on source spectrum,intensity, and sky position. Super-AGILE operating inthe 15–45 keV band has a spatial resolution of 6 arcmin(pixel size). This translates into a positional accuracy of1–3 arcmins for relatively strong transients at the Crabflux level.
3.2. Large FOV monitoring of gamma-ray sources
A crucial feature of AGILE is its large FOV for both thegamma-ray and hard X-ray detectors. Fig. 10 shows typicalgamma-ray FOVs obtained for a sequence of AGILEpointings. Relatively bright AGNs and galactic sourcesflaring in the gamma-ray energy range above a flux of10�6 ph=cm2=s can be detected within a few days by theAGILE Quicklook Analysis. We conservatively estimatethat for a 2-year mission AGILE is potentially able todetect hundreds of gamma-ray flaring AGNs and othertransients. The very large FOV will also favor the detectionof GRBs above 30MeV. Taking into account the high-energy distribution of GRB emission above 30MeV, weconservatively estimate that �1 GRB/month can bedetected and imaged in the gamma-ray range by theGRID. Super-AGILE may be able to detect about 30% ofthe sources detected by INTEGRAL [14]; about 10 hardX-ray sources per day are expected to be detected by SAfor typical pointings of the galactic plane.
ARTICLE IN PRESS
Fig. 10. An example of a sequence of AGILE pointings showing the very
large gamma-ray FOV (in galactic coordinates). Typical AGILE pointings
will last 3 weeks and are subject to solar panel constraints.
M. Tavani et al. / Nuclear Instruments and Methods in Physics Research A 588 (2008) 52–6258
3.3. Fast reaction to strong high-energy transients
The existence of a large number of variable gamma-raysources (extragalactic and near the galactic plane, e.g.,Ref. [15]) makes necessary a reliable program for quickresponse to transient gamma-ray emission. QuicklookAnalysis of gamma-ray data is a crucial task to be carriedout by the AGILE Team. Prompt communication ofgamma-ray transients (requiring typically 1–3 days tobe detected with high confidence for sources above10�6 ph=cm2=s) will be ensured. Detection of short time-scale (seconds/minutes/hours) transients (GRBs, SGRs,and other bursting events) is possible in the gamma-rayrange. A primary responsibility of the AGILE Team will beto provide positioning of short-timescale transient asaccurate as possible, and to alert the community thoughdedicated channels.
3.4. Accumulating exposure on galactic and extragalactic
sky areas
The AGILE average exposure per source will be largerby a factor of �4 for a 1-year sky-survey programcompared to the typical exposure obtainable by EGRETfor the same time period. Deep exposures for selectedregions of the sky can be obtained by a proper programwith repeated overlapping pointings. This can be particu-larly useful to study selected galactic and extragalacticsources.
3.5. High-precision timing
AGILE detectors have optimal timing capabilities.The on-board GPS system allows to reach an absolutetime tagging precision for individual photons better than2ms. Depending on the event characteristics, absolutetime tagging can achieve values near 1–2ms for the
Silicon-Tracker, and 3–4ms for the Mini-Calorimeter andSuper-AGILE.Instrumental deadtimes will be unprecedentedly small
for gamma-ray detection. The GRID deadtime will belower than 200ms (improving by almost three orders ofmagnitude the performance of previous spark-chamberdetectors such as EGRET). Taking into account thesegmentation of the electronic readout of MCAL andSuper-AGILE detectors (30 MCAL elements and 16Super-AGILE elements) the MCAL and SA effectivedeadtimes will be less than those for individual units. Weobtain �2ms for MCAL, and 5ms for SA. Furthermore,a special memory will ensure that MCAL events detectedduring the Si-Tracker readout deadtime will be auto-matically stored in the GRID event. For these events,precise timing and detection in the �1–200MeV range canbe achieved with temporal resolution well below 100ms.This may be crucial for AGILE high-precision timinginvestigations.
3.6. AGILE and GLAST
AGILE and GLAST [16] are complementary missions inmany respects: we briefly outline here some importantpoints, postponing a more general discussion [17].The GLAST gamma-ray instrument covers a broadspectrum and is especially optimized in the high-energyrange above 1GeV. On the other contrary, AGILEis optimized in the range below 1GeV with emphasis tothe simultaneous hard X-ray/gamma-ray imaging witharcminute angular resolution. The GLAST large gamma-ray effective area allows deep pointings and good imagingwithin a few arcminutes for strong gamma-ray sources.AGILE has a gamma-ray effective area near 100MeVsmaller by a factor of �4 compared to the upper portionof the LAT instrument on board of GLAST. However,it can reach arcminute positioning of sources becauseof Super-AGILE. Furthermore, the GLAST sky-scanningmode adopted during the first phase of the mission,and the AGILE fixed pointing strategy offer a wayfor joint investigations. In the overlapping pointedregions, strong time variability of gamma-ray sources(1–2 day timescale) can be very effectively studiedsimultaneously by the two missions. AGILE with itshard X-ray imager can also provide additional and veryuseful information.The most relevant feature of AGILE is its unique
combination of a large-FOV hard X-ray imager togetherwith a gamma-ray imager. AGILE will be able to reacharcminute positioning of sources emitting in the hardX-ray range above 10mCrab. Furthermore, AGILE canreact to transients, and can point at gamma-ray sourcesdetected by other missions to position the source within theFOV of both the GRID and SA. Interesting gamma-raytransients detected by GLAST might then be pointed byAGILE for a broad-band study of their temporal andspectral properties.
ARTICLE IN PRESSM. Tavani et al. / Nuclear Instruments and Methods in Physics Research A 588 (2008) 52–62 59
4. Scientific objectives
Currently, nearly 270 gamma-ray sources above 30MeVwere detected (with only a small fraction, 30%, identifiedas AGNs or isolated pulsars) [18–21]. AGILE fits into thediscovery path followed by previous gamma-ray missions(SAS-2, COS-B, and EGRET) and be complementary toGLAST. We summarize here the main AGILE’s scientificobjectives.
�
Active galactic nuclei: For the first time, simultaneousmonitoring of a large number of AGNs per pointing willbe possible. Several outstanding issues concerning themechanism of AGN gamma-ray production and activitycan be addressed by AGILE including: (1) the study oftransient vs. low-level gamma-ray emission and duty-cycles [22]; (2) the relationship between the gamma-rayvariability and the radio-optical-X-ray-TeV emission;(3) the correlation between relativistic radio plasmoidejections and gamma-ray flares; (4) hard X-ray/gamma-ray correlations. � GRBs: A few GRBs were detected by the EGRET sparkchamber [23]. This number appears to be limited by theEGRET FOV and sensitivity and probably not by theintrinsic GRB emission mechanism. GRB detection rateby the AGILE-GRID is expected to be at least a factorof several larger than that of EGRET, i.e., X2–5 events/year). The small GRID deadtime (�500 times smallerthan that of EGRET) allows a better study of the initialphase of GRB pulses (for which EGRET response wasin many cases inadequate). AGILE is expected to beefficient in detecting photons above 10GeV because oflimited backsplashing. Super-AGILE will be able tolocate GRBs within a few arcminutes, and will system-atically study the interplay between hard X-ray andgamma-ray emissions. Special emphasis will be given tothe search for sub-millisecond GRB pulses indepen-dently detectable by the Si-Tracker, MCAL and Super-AGILE. One of the early results by AGILE has been thedetection of several GRBs. Fig. 12 shows the on-boarddetection by Super-AGILE of GRB 070724B and itslocalization (Fig. 13) [24]. � Diffuse galactic and extragalactic emission: The AGILEgood angular resolution and large average exposure willfurther improve our knowledge of cosmic ray origin,propagation, interaction and emission processes. Wealso note that a joint study of gamma-ray emission fromMeV to TeV energies is possible by special programsinvolving AGILE and new-generation TeV observa-tories of improved angular resolution. � Gamma-ray pulsars: AGILE will contribute to the studyof gamma-ray pulsars (PSRs) in several ways: (1)improving timing and lightcurves of known gamma-ray PSRs; (2) improving photon statistics for gamma-ray period searches; (3) studying unpulsed gamma-rayemission from plerions in supernova remnants andstudying pulsar wind/nebula interactions, e.g., as in thegalactic sources recently discovered in the TeV range[25]. Particularly interesting for AGILE are the �30 newyoung PSRs discovered [26] in the galactic plane by theParkes survey.
� Search for non-blazar gamma-ray variable sources in thegalactic plane, currently a new class of unidentifiedgamma-ray sources such as GRO J1838-04 [15].
� Galactic sources, micro-quasars, new transients: A largenumber of gamma-ray sources near the galactic planeare unidentified, and sources such as 2CG 135þ 1=LS I61 þ61 303 can be monitored on timescales of months.Cyg X-1 will be also monitored and gamma-rayemission above 30MeV will be intensively searched.Galactic X-ray jet sources (such as Cyg X-3, GRS1915þ 10, GRO J1655-40 and others) can producedetectable gamma-ray emission for favorable jet geo-metries, and a TOO program is planned to follow-upnew discoveries of micro-quasars. � Fundamental physics: quantum gravity: AGILE detectorsare suited for quantum gravity studies [10]. Theexistence of sub-millisecond GRB pulses lasting hun-dreds of microseconds [27] opens the way to study QGdelay propagation effects by AGILE detectors. Particu-larly important is the AGILE Mini-Calorimeter withindependent readout for each of the 30 CsI bars of smalldeadtime (�20ms) and absolute timing resolution(�3ms). Energy dependent time delays near �100msfor ultra-short GRB pulses in the energy range0.3–3MeV can be detected. If these GRB ultra-shortpulses originate at cosmological distances, sensitivity tothe Planck’s mass can be reached [10].5. The mission
The AGILE spacecraft is of the MITA class and isdeveloped by Carlo Gavazzi Space (CGS) as primecontractor and by Oerlikon-Contraves. The spacecraftprovides a 3-axis stabilization with an accuracy near0.5–11. The final satellite pointing reconstruction isrequired to reach an accuracy of �1 arcmin by a set oftwo Star Sensors. A GPS transceiver will also ensure an on-board timing accuracy within 2ms. The AGILE scientificinstrument generates under normal conditions a telemetryrate of �50 kbit=s. The satellite downlink telemetry rate is512 kbit=s, that is adequate to transmit at every passageover the ground station all the satellite and scientific data.The fixed solar panels configuration and the necessity to
have them always exposed to the Sun imposes someconstraints on the AGILE pointing strategy. However, inpractice the AGILE large FOV does not sensibly limit theaccessible sky: only the solar and anti-solar directions areexcluded from direct pointings. The AGILE Pointing Planis being finalized and will be ready in advance before thestart of Cycle-1 (first year). AGILE might react to transientevents of great importance occurring outside the accessibleFOV. For transients detected by the AGILE-GRID and
ARTICLE IN PRESSM. Tavani et al. / Nuclear Instruments and Methods in Physics Research A 588 (2008) 52–6260
not by Super-AGILE, a minor re-pointing (20–301) isenvisioned to allow the coverage of the gamma-raytransient also by the X-ray imager within 1 day. A drasticre-pointing strategy (Target-of-Opportunity, TOO) is fore-seen for events of major scientific relevance detected byother observatories.
The AGILE orbit is quasi equatorial, with an inclina-tion of 2.51 from the Equator and average altitude of540 km. A low earth orbit (LEO) of small inclinationis a clear plus of the mission because of the reducedparticle background as verified in orbit by all instru-ment detectors. Also a low-inclination orbit optmizes theuse of the ASI communication ground base at Malindi(Kenya).
6. The AGILE science program
AGILE is a small scientific mission with a scienceprogram open to the international scientific community.The AGILE Mission Board (AMB) oversees the scientificprogram, determines the pointing strategy, and authorizesTOO observations in case of exceptional transients. Asubstantial fraction of the gamma-ray data will be availablefor the AGILE Guest Observer Program (GOP) that willbe open to the international community on a competitivebasis. The AGILE Cycle-1 GOP is scheduled to start inDecember, 2007.
6.1. Data analysis and scientific ground segment
AGILE science data (about 300Mbit/orbit) are teleme-tered from the satellite to the ASI ground station inMalindi (Kenya) at every satellite passage (about 90min).A fast ASINET connection between Malindi and theTelespazio Satellite Control Center at Fucino and thenbetween Fucino and the ASI Science Data Center (ASDC)ensures the data transmission every orbit. The AGILEMission Operations Center is located at Fucino and will beoperated by Telespazio with scientific and programmaticinput by ASI and the AGILE Science Team through theASDC.
Scientific data storage, quicklook analysis and theGOP will be carried out at ASDC. After pre-processing,scientific data (Level-1) will be corrected for satelliteattitude data and processed by dedicated software pro-duced by the AGILE Team in collaboration with ASDCpersonnel. Background rejection and photon list determi-nation are the main outputs of this first stage of processing.Level-2 data will be at this point available for a fullscientific analysis. Gamma-ray data generated by theGRID will be analyzed by special software producing: (1)sky-maps, (2) energy spectra, (3) exposure, (4) point-sourceanalysis products, and (5) diffuse gamma-ray emission.This software is aimed to allow the user to perform acomplete science analysis of specific pointlike gamma-raysources or candidates. This software will be available forthe GOP. Super-AGILE data will be deconvolved and
processed to produce 2-D sky images through a correlationof current and archival data of hard X-ray sources. GRBdata will activate dedicated software producing lightcurves,spectra and positioning both in the hard X-ray (18–60 keV)and gamma-ray energy (30MeV–30GeV) ranges. TheAGILE data processing goals can be summarized asfollows:
�
Quicklook Analysis (QA) of all gamma-ray and hardX-ray data, aimed at a fast scientific processing (within afew hours/1 day depending on source intensity) of allAGILE science data; � web-availability of QA results to the internationalcommunity for alerts and rapid follow-up observations;
� GRB positioning and alerts through the AGILE FastLink, capable of producing alerts within 1–2min sincethe event;
� standard science analysis of specific gamma-ray sourcesopen to a GOP;
� web-availability of the standard analysis results of thehard X-ray monitoring program by Super-AGILE.
7. In-orbit calibration and early results
After the successful launch, the AGILE satellite carriedout a 2-month phase of functional checkout tests (Com-missioning phase). The satellite and all instrument detec-tors were turned on according to a nominalCommissioning Plan and subsequently tested for acquisi-tion of scientific data. All detectors behaved nominallywith a quite good measured particle background. The startof the nominal gamma-ray and X-ray imaging dataacquisition was carried out in May and early June,respectively. Preliminary gamma-ray data of the Velapulsar obtained in early June, 2007 confirmed the goodquality of the background rejection and the imagingcapability of the GRID. The scientific operations startedin early July, 2007 with a long 2-month observation of theVela PSR region. At the end of August, 2007, AGILEdevoted about 1 week to a preliminary pointing of theGalactic Center region. Finally, AGILE carried out thegamma-ray and hard X-ray calibration with the Crabpulsar during the months of September and October, 2007.Fig. 11 shows a 1-day integration of the gamma-ray sky ofthe Crab and Geminga pulsar region. This field wasrepeatedly observed with 1-day pointings with the Crabpulsar position at different angles with respect to theInstrument axis. The in-orbit calibration phase wassuccessfully completed at the end of October, 2007achieving a good performance for both the gamma-rayand hard X-ray imagers.The nominal Science Verification Phase observations
were interrupted for specific satellite special pointings thatwere carried out either for planned multifrequencycampaigns (for 3C 279 and 3C 273 in early July, 2007)or reacting to external alerts for relevant optical activity
ARTICLE IN PRESS
Fig. 11. AGILE 1-day gamma-ray counts map for photons above
100MeV of the Crab Nebula and Geminga region.
12001000
800600400200
0-200
coun
ts p
er b
in
12001000
800600400200
0-200
coun
ts p
er b
in
-60 -40 -20 0 20 40 60�x [degree]
-60 -40 -20 0 20 40 60�x [degree]
Fig. 12. The Super-AGILE detection of GRB 070724B in the energy
range 20–60keV [24]. Deconvolved sky images of the X and Z one-
dimensional detectors.
0.5 1 1.5 2
14m00.0s 12m00.0s 1h10m00.0s 08m00.0s
Swift/XRT29m59.8s
57d44m59.8s
59m59.8sSuperAGILE
x
Fig. 13. The Super-AGILE localization of GRB 070724B together with
the Swift/XRT image. Early (blue circle) and refined (blue box, 60 � 60Þ
Super-AGILE positioning superimposed with the X-ray image by Swift
showing the afterglow.
M. Tavani et al. / Nuclear Instruments and Methods in Physics Research A 588 (2008) 52–62 61
of blazars (3C 454.3 in mid-July, 2007; TXS 0716þ 714at the end of October, 2007). At the time of this writing(early December, 2007) AGILE announced the gamma-ray detection of several blazars (e.g., ATEL nos. 1278,1300) and of a gamma-ray transient in the Cygnusregion (ATEL nos. 1308, 1320). Several GRBs weredetected by the Super-AGILE and the MCAL detectors.Figs. 12 and 13 show the Super-AGILE detection andpositioning of GRB 070724B [24]. On December 1, 2007,the AGILE Cycle-1 program of nominal scientific observa-tions started.
7.1. Multiwavelength observations program
The scientific impact of a high-energy mission such asAGILE (broad-band energy coverage, very large FOV) isgreatly increased if an efficient program for fast follow-upand/or monitoring observations by ground-based andspace instruments is carried out. The AGILE ScienceProgram overlaps and be complementary to those of manyother high-energy space missions (INTEGRAL, XMM-Newton, Chandra, SWIFT, Suzaku, GLAST) and ground-based instrumentation.The AGILE Science Program will involve a large
astronomy and astrophysics community and emphasizesa quick reaction to transients and a rapid communicationof crucial data. Past experience shows that in manyoccasions there was no fast reaction to gamma-raytransients (within a few hours/days for unidentifiedgamma-ray sources) that could not be identified. AGILEwill take advantage, in a crucial way, of the combination ofits gamma-ray and hard X-ray imagers.An AGILE Science Group (ASG) aims at favoring
the scientific collaboration between the AGILE Teamand the community especially for coordinating multi-wavelength observations based on AGILE detectionsand alerts. The ASG is open to the internationalastrophysics community and consists of the AGILETeam and qualified researchers contributing with theirdata and expertise in optimizing the scientific return ofthe mission. Several working groups are operationalon a variety of scientific topics including blazars, GRBs,pulsars, and galactic compact objects. The AGILE Team isalso open to collaborations with individual observinggroups.
ARTICLE IN PRESSM. Tavani et al. / Nuclear Instruments and Methods in Physics Research A 588 (2008) 52–6262
Updated documentation on the AGILE Mission can befound at the web sites http://asdc.asi.it, and http://agile.iasf-roma.inaf.it.
Acknowledgments
The AGILE program is developed under the auspices ofthe Italian Space Agency with co-participation by theItalian Institute of Astrophysics (INAF) and by the ItalianInstitute of Nuclear Physics (INFN). Research partiallysupported under the Grants ASI-I/R/045/04 and ASI-I/089/06/0. We acknowledge the crucial programmaticsupport of the ASI Directors of the Observation of the
Universe Unit (S. Di Pippo) and of the Small mission Unit(F. Viola, G. Guarrera), as well as of the AGILE MissionDirector L. Salotti.
References
[1] M. Tavani, G. Barbiellini, P. Caraveo, S. Di Pippo, M. Longo,
S. Mereghetti, A. Morselli, A. Pellizzoni, P. Picozza, S. Severoni,
F. Tavecchio, S. Vercellone, AGILE Phase A Report, 1998.
[2] M. Tavani, et al., in: M. McConnell (Ed.), Proceedings of the Fifth
Compton Symposium AIP Conference Proceedings, vol. 510, 2000,
p. 746.
[3] M. Tavani, Texas in Tuscany, XXI Symposium on Relativistic
Astrophysics, Florence, Italy, 9–13 December 2002, 2003, p. 183.
[4] A. Morselli, et al., Nucl. Phys. B 85 (1–3) (2000) 22.
[5] A. Bakaldin, A. Morselli, P. Picozza, et al., Astropart. Phys. 8 (1997)
109.
[6] G. Barbiellini, A. Morselli, et al., Nucl. Phys. B 43 (1995) 253.
[7] G. Barbiellini, A. Morselli, et al., Nucl. Instr. and Meth. 354 (1995) 547.
[8] G. Barbiellini, et al., SPIE 2478 (1995) 239.
[9] R.L. Golden, A. Morselli, P. Picozza, et al., Il Nuovo Cimento B 105
(1990) 191.
[10] M. Tavani, 2008, in preparation.
[11] M. Tavani, et al., The AGILE Scientific Instrument, Nucl. Instr. and
Meth. (2007), to be submitted.
[12] G. Barbiellini, et al., Nucl. Instr. and Meth. A 490 (2002) 146.
[13] M. Feroci, et al., Nucl. Instr. and Meth. A 581 (2007) 728.
[14] A.J. Bird, et al., Astrophys. J 636 (2006) 765.
[15] M. Tavani, et al., Astrophys. J. 479 (1997) L109.
[16] hhttp://glast.gsfc.nasa.govi, hhtpp://glast.stanford.edui.
[17] M. Tavani, et al., The AGILE Mission and its Scientific Program,
2008, in preparation.
[18] R.C. Hartman, et al., Astrophys. J. Sci. 123 (1999) 279.
[19] D.J. Thompson, et al., Astrophys. J. Sci. 101 (1995) 259.
[20] D.J. Thompson, et al., Astrophys. J. Sci. 107 (1996) 227.
[21] D.J. Thompson, et al., in: Proceedings of the Fourth CGRO
Symposium, AIP Conference Series, vol. 410, 1998, p. 39.
[22] S. Vercellone, et al., Mon. Not. R. Astron. Soc. 353 (2004) 890.
[23] E.J. Schneid, et al., in: AIP Conference Proceedings, vol. 384, 1996,
p. 253.
[24] E. Del Monte, et al., Astron. Astrophys. 478 (2008) L5.
[25] F. Aharonian, et al., Astrophys. J. 636 (2006) 777.
[26] M. Kramer, et al., Mon. Not. R. Astron. Soc. 342 (2003) 1299.
[27] C.L. Bhat, et al., Nature 359 (1992) 217.
