The Gamma Ray Large Area Space Telescope (GLAST) Tsunefumi Mizuno, 1 - - PDF document

Astronomy and Astrophysics of Extreme Universe 1

The Gamma Ray Large Area Space Telescope (GLAST)

Tsunefumi Mizuno,1 on behalf of the GLAST LAT team

(1) Hiroshima University, 1-3-1 Kagamiyama, Higashi-Hiroshima, Hiroshima, 739-8526, Japan

Abstract The Gamma Ray Large Area Space Telescope (GLAST) consists of two instru- ments, Large Area Telescope (LAT) and GLAST Burst Monitor (GBM), and is being developed by an international collaboration. The main instrument LAT is a pair-conversion gamma-ray telescope which observes gamma-ray sky with wide field-of-view (∼ 2.4 sr) and high sensitivity in ∼ 20 MeV–300 GeV. The secondary instrument GBM covers energy range below 30 MeV and monitor gamma-ray bursts with even wider field-of-view (∼ 9 sr). The GLAST will survey gamma-ray sky and provide vital information on almost all kinds of gamma-ray astronomical

• sources. The GLAST is scheduled to be launched in December 2007.

1. GLAST Mission Overview The Gamma Ray Large Area Space Telescope (GLAST) is a next generation high-energy gamma-ray observatory developed under an international collabora- tion among United States, Japan and European countries. The main instrument, the Large Area Telescope (LAT), follows the footsteps of the CGRO-EGRET ex- periment [1] [2] and will provide unprecedented sensitivity to gamma-ray sources in the energy range of ∼ 20 MeV–300 GeV. The GLAST Burst Monitor (GBM), a successor to BATSE experiment on-board CGRO, was selected as a complemen- tary instrument for the GLAST LAT and will monitor gamma-ray bursts (GRBs) in the energies between ∼ 10 keV to ∼ 30 MeV. The GLAST satellite will be launched in late 2007 or early 2008 by a Delta-II rocket from Kennedy Space Center in Florida. 2. Instrumentation 2.1. Large Area Telescope Overview The GLAST LAT [3] has been developed and built by an international col- laboration among United States, Japan, France, Italy and Sweden. The LAT is designed to measure the direction of gamma-rays incident over a wide energy range of ∼ 20 MeV–300 GeV and a wide field-of-view (FoV) of ∼ 2.4 sr, while rejecting background from cosmic-rays. The LAT is a pair-conversion telescope with a Si-strip tracker (TKR) and a CsI hodoscopic array of calorimeter (CAL), each consisting of a 4 x 4 array of 16 modules called ”towers”. A segmented anti- coincidence detector shield (ACD) made of plastic scintillators covers the tracker array, and data acquisition system (DAQ) utilizes prompt signals from the TKR, CAL and ACD subsystems to issue a trigger. Upon triggering, the DAQ initiates the read-out of three subsystems and utilizes on-board event processing to reduce number of cosmic-ray events to fit to the band width of available down-link. A prototype single tower had been tested and validated through a series of beam tests and a balloon flight [4] [5] [6] [7]. The LAT instruments is schematically shown in Figure 2.1..

c 2007 by Universal Academy Press, Inc.

2

• Fig. 1.

A schematic drawing of the GLAST LAT composed

• f

the TKR, CAL and ACD subsystems. Note that the instrumental design has been updated (e.g., converters

• f flight model TKR are tungsten

instead of Lead).

Tracker Each TKR module has 18 tracking planes, each consisting of two layers (to measure x and y direction) of single-sided 400 µm-thick and 228 µm-pitch silicon strip detectors [8] and tungsten converters. The support structure for the detec- tors and converter foils is composed of a stack of 19 composite panels (trays) of about 3 cm thickness made of carbon-composite assembly. All trays are of similar construction, although the top and bottom ones are special, with detectors on

• nly a single face. An x,y measurement plane consists of a layer of detectors on

the bottom of one tray together with an orthogonal detector layer on the top of the tray just below, with a 2 mm separation. The tungsten converter foils lie immediately above the upper detector layer. The strips on the top and bottom of a given tray are parallel, while alternate trays are rotated 90◦ with respect to each

• ther. There are 16 x,y planes at the top of the tracker with converter foils, among

which upper 12 converters are thinner (3% radiation length) to maintain good an- gular resolution, and next four converters are thicker (18% radiation length) to achieve high effective area within a limited size of the TKR module. The lowest two x,y planes have no converter. Thanks to its fine position resolution (228 µm), silicon-strip TKR has much improved position accuracy and angular resolution than those of EGRET. It also makes it possible for the LAT to have much larger effective area, wider FoV and smaller dead time. Details of the GLAST TKR can be found in [9]. Calorimeter Each CAL module consists of 96 CsI(Tl) scintillators, with each crystal of size 2.7 cm (width) × 2.0 cm (height) × 32.6 cm (length). The crystals are optically isolated from each other and are arranged horizontally in 8 layers of 12 crystals each, giving the total depth of 8.6 radiation lengths (for a total instrument depth

• f 10.1 radiation lengths).

Each calorimeter module layer is aligned 90◦ with respect to its neighbors, forming an x and y (hodoscopic) array. The segmentation

3

allows spatial imaging of the electromagnetic shower and accurate reconstruction

• f its direction, because each CsI crystal provides three spatial coordinates for

the energy deposited within: two discrete coordinates from the physical location

• f the crystal and the third coordinate determined by measuring the light yield

asymmetry at the ends of the crystal along its long dimension. The calorimeter’s shower imaging capability contributes significantly to the background rejection, and enables the shower leakage correction which results in the high-energy reach

• f the LAT up to ∼ 300 GeV. Thanks to its large collection area, the CAL also

works as a highly efficient cosmic electron detector (e.g., [10] [11]). For more detail

• f the CAL, refer to [12].

Anticoincidence detector The role of the ACD is to provide charged particle background rejection; there- fore its main requirement is to have very high efficiency (of 0.9997 when averaged

• ver the whole area) for detection of singly-charged particles entering the tracking

detector from the top or side of the LAT. The ACD veto signal on-board reduces the trigger rate to a level compatible to the data transmission rate to the ground. The ACD data will also be used during off-line analysis to achieve an ultimate background rejection efficiency. In order for the LAT to measure gamma-rays with energies up to 300 GeV, one has to take care of a problem called backsplach effect: a small fraction of secondary particles (mostly 100-1000 keV photons) from the electromagnetic shower created by the incident high energy photon in the CAL travel backward through the tracker and create veto signals in the ACD. This effect was present in the EGRET experiment and limited the sensitivity at ∼ 10 GeV. To minimize these false vetos and maintain sufficient sensitivity for gamma-rays above 100 GeV, the LAT ACD is segmented into 89 tiles and only ACD segments in the projected path of the incident photon is considered for vetoing, thereby dramatically reducing the area of ACD that can contribute to backsplash. See [13] for more details of the ACD instrumentation and testing. Expected performance A combination of three sub-systems provides us with an excellent performance

• f the LAT and unprecedented sensitivity for gamma-ray objects. The key pa-

rameters of the LAT performance are summarized in Figure 2. and Table 1.. A better angular resolution and larger effective area (compared to those of EGRET) result in much improved sensitivity for point sources and diffuse emission. A much larger FoV allows us to monitor gamma-ray sky and search for transient sources and flare of known objects continuously. A better timing resolution is a great advantage in searching for gamma-ray pulsars. 2.2. GLAST Burst Monitor Aside from the main instrument LAT, a secondary instrument GBM (GLAST Burst Monitor) is equipped to the GLAST observatory. The GBM is designed to be complementary to the LAT: it has even wider FoV (∼ 9 sr) and sensitivity for low-energy photons down to ∼ 10 keV in order to increase the detection rate

• f the gamma-ray bursts (GRBs) and transients, and extend the energy coverage

for the spectroscopic study. The GBM was developed by a collaboration between United States and Germany. Like CGRO-BATSE, the GBM design is based on the use of two types cylindrical scintillation detectors. One is an array of 12 sodium iodide (NaI) detectors having sensitivity in the energy range of ∼ 10 keV–1 MeV covering a typical spectral

4 Fig. 2. Expected performance

• f

the GLAST LAT. Angular resolution, ef- fective area and relative effective area as a function

• f

incident angle (i.e., FoV) are given by left, middle and right panel, respectively. See http://www-glast.slac.stanford.edu/software/IS/glast_lat_performance.htm for updated information. Table 1. GLAST-LAT expected performance based on computer simulation and com- parison with EGRET

GLAST LAT EGRET Energy Range 20 MeV–300 GeV 20 MeV–30 GeV Energy Resolution ≤ 10% 10% Effective area ≥ 8000 cm−2 1500 cm−2 Field of view ≥ 2.4 sr 0.5 sr Angular Resolution 3.5◦@100 MeV/0.15◦@10 GeV 5.8◦@100 MeV Sensitivity (≥ 100 MeV) 4 × 10−9 ph s−1 cm−2 10−7 ph s−1 cm−2 Deadtime ≤ 100 µs 100 ms

5

break of GRBs. The other detector is composed of two bismuth germanate (BGO) scintillators to provide a spectral overlap with the LAT. Energy range covered by BGOs is from ∼ 150 keV to ∼ 30 MeV. As a result, the GLAST will provide us with an extremely broad energy coverage up to 6 orders of magnitude. An example

• f the spectrum expected to be measured by the GBM and the LAT is shown by

Figure 3.. The accuracy of the burst location is estimated to be 15◦ by a data processing on board and will be improved down to 5◦ after the transmission of data to the ground. The GBM will also be used to re-point the LAT at particularly interesting bursts for performing afterglow observations. Current status of the GBM can be found in [14] and references therein.

• Fig. 3.

Simulation of the spec- trum of GRB940217 as would be observed by the GBM and the LAT [15]. For the GBM, data points from the NaI and BGO detectors are indicated separately.

3. Ititial Operation After the initial checkout of on-orbit data for ∼ 60 days after the launch (Phase 0), the GLAST will be operated in a full sky survey mode for the first year of scientific operation (Phase 1). The photon lists will be used by the instrument team for a detailed instrument calibration and key projects such as a catalog

• generation. During this period, all GBM data will be released, and high level

LAT data (time-resolved spectra) of ∼ 20 selected sources and all transients will be posted on the web regularly. Each photon list will become public about a month after Phase 1 ends. The rest of the GLAST mission is defined as Phase 2. During this period, the GLAST data will be available to the scientific community as soon as the data processing has been completed. Default operation mode is a sky survey mode but observations driven by guest observer proposal selected by peer review are also possible. Since the LAT has a very large FoV and will scan the sky, there is no proprietary period of data even for guest observer programs. The number of sources of interest is expected to be increased as the data is accumulated during the survey. Current list of LAT-monitoring sources are given in Table 2.. For more detail of the operation and data policy, see http://glast.gsfc.nasa. gov/ssc/data/policy/. 4. Science with GLAST The key science objectives addressed by the GLAST LAT are largely motivated by the discoveries of EGRET (30 MeV – 10 GeV) and of ground-based atmo- spheric Cherenkov telescopes (≥ 100 GeV). Thanks to the combined advances in point spread function, effective area, energy range and the FoV, the LAT will have a point-source sensitivity of ∼ 4 × 10−9 ph s−1 cm−2 (E ≥ 100 MeV), about 30 times higher than that of EGRET. The GLAST GBM, the successor of CGRO- BATSE, is expected to find ∼ 200 burst per year thanks to its very large FoV (9 sr) and a wide energy coverage (10 keV–30 MeV). About 1/3 of them will be inside LAT FoV and allow us to conduct a wide-band spectroscopic study (from 10 keV up to 100 GeV) of bright bursts. The science topics pursued by GLAST includes particle acceleration and high-energy radiation mechanism in the pulsar

6 Table 2. GLAST LAT monitoring sources as of May 2007. Know TeV sources are indicated with underline. type source names Blazar 0208-512, 0235+164, PKS0528+134, PKS 0716+714, 0827+243, OJ 287 Mrk 421, WCOM 1259+285, 3C 273, 3C 279, 1406-076, H 1426+428 1510-089, PKS 1622-297, 1633+383, Mrk 501, 1730-130 NRAO 530 1ES 1959+650, PKS 2155-304, BL Lacertae (2200+420), 3C 454.3 1ES 2344+514 HMXB LSI+61 303 Transients any if flux exceeds 2 × 10−6 ph s−1 cm−2

magnetosphere, composition and mechanism of jets seen in micro-quasars and ac- tive galactic nuclei, study of supernova remnants as an origin of cosmic-rays, high energy behavior of GRBs, identification of EGRET unidentified sources and dis- covery of new classes of gamma-ray objects, matter and cosmic-ray distribution in

• ur galaxy, production of cosmic-rays in galaxies and cluster of galaxies, resolving

extragalactic diffuse gamma-ray emission into each object and probing the na- ture of Dark Matter. Thanks to the high sensitivity of the GLAST, population studies of gamma-ray pulsars, blazars and GRBs are also possible. In addition, the GLAST-LAT will be the first instrument to achieve hourly to daily variability

• f the entire Universe in the gamma-ray band. We expect many serendipitous

discovery in this unexplored temporal domain. In this section, we briefly describe a couple of scientific topics that will be progressed by the GLAST. 4.1. Pulsars and Pulsar Wind Nebulae Pulsars, with their unique temporal signature, were the only EGRET population

• f Galactic point sources definitively identified by other wavebands. There were

seven gamma-ray pulsars detected by EGRET. With the improved sensitivity, the GLAST LAT will detect more than 100 new gamma-ray pulsars and allow us to conduct an evolution study of pulsars. Surrounding young pulsars are bright non-thermal pulsar wind nebulae (PWNe), where X-ray and gamma-ray emission is generated through the interaction of accelerated particles with circumstellar magnetic fields and photons. EGRET has already detected a clear signature of PWN gamma-ray emission from the Crab pulsar [16], and TeV gamma-rays via Compton scattering have been found from several PWNe (e.g., [17] and references therein). The connection of the synchrotron spectrum in GeV and the inverse Compton flux in TeV, provided by a joint observation of the GLAST LAT and ground atmospheric Cherenkov telescopes, is valuable in constraining the magnetic field and injected particle spectrum. One basic question not answered yet is the mechanism of the the high energy emission from rotation-powered pulsars. Two completing models have been pro- posed, one is a polar cap model [18] which predicts the high energy emission arises near the neutron star surface and the other is a outer gap model [19] according to which non-thermal electromagnetic spectrum is generated at a significant fraction

• f the light cylinder distance. In addition to geometrical (beam-shape) differences,

two models predict substantially different high-energy cutoff in pulsar spectrum, and can be distinguished by the high sensitivity observation by the GLAST LAT as shown by Figure 4..

7 Fig. 4. Modeled high energy spectrum of the Vela pulsar

• btained by a one-year obser-

vation by the GLAST LAT [20]. Predictions of the po- lar cap model and the outer gap model are shown by green and blue line/data, respec-

• tively. EGRET data in black

crosses is overlayed.

4.2. Supernova Remnants The origin and acceleration mechanism of cosmic rays is one of the biggest mys- teries since their discovery almost a hundred years ago. Cosmic-rays up to 1015 eV (the knee energy) have long been thought to be shock-accelerated at the rim of supernova remnants (SNRs). Discoveries of non-thermal X-ray emission from rims

• f SNRs (e.g., [21]) provide a firm evidence of electrons accelerated to TeV energies

for the first time. In addition, recently the H.E.S.S. experiment succeeds in de- tecting TeV gamma-rays from Galactic SNRs (e.g., [22] [23]). However, the origin

• f accelerated particles responsible to TeV emission is still uncertain. Although

the hadronic scenario (gamma-rays from decay of π0 generated via hadronic in- teraction) is favored, leptonic scenario (TeV gamma-rays are generated through inverse Compton scattering by accelerated electrons) has not been ruled out yet. The GLAST LAT has the spatial and spectral sensitivity to resolve this question and thus will constrain the origin of cosmic rays. Two models predicts signifi- cantly different spectrum in GeV energy band covered by the LAT and can be distinguished unambiguously, as illustrated by Figure 5..

• Fig. 5.

Spectral energy distribu- tion of the shell-type SNR RX J1713.7-3946 above 100 MeV [24]. The black data points are measurements with H.E.S.S., and the blue and red data points corresponds to sim- ulated LAT data assuming hadronic scenario and leptonic scenario, respectively.

8

4.3. Active Galactic Nuclei Before EGRET, 3C 273 was the only active galactic nucleus (AGN) known to emit high energy gamma-rays. Unexpected finding by EGRET is that Blazar AGN is the largest population of gamma-ray sources in GeV energy range: more than 60 Blazars are identified with high confidence [25]. The broadband spec- tra of blazars show two pronounced peaks (e.g., [26]). One is a lower-frequency component peaking between radio and X-ray energies believed to be synchrotron emission from energetic non-thermal electrons in blazar jets, and the other is a high-frequency component peaking in gamma-rays thought to be generated via inverse Compton scattering of soft photons by high energy electrons. Therefore multi-wavelength observation is essential to study jets in blazars. Extrapolation

• f the EGRET LogN-LogS relation indicates that the LAT will detect several

thousand AGNs and enables detailed population studies and the investigation of the cosmological evolution of central massive black holes. High sensitivity and wide FoV of the LAT will allow blazar variability to be monitored continuously. Flares as bright as that observed by EGRET from 3C 279 [27] will be measurable with LAT on time scales less than a day (Figure 6.). Simultaneous multi-wavelength observations of such strong flares led by GLAST LAT and detailed comparison with the theoretical prediction in time scale of hours will constrain the composition (i.e., hadronic or leptonic model of accelerated particles) and mechanism (e.g., synchrotron self-Compton or external Compton scattering scenario to produce a Compton peak) of jets. In addition, the redshift dependence of spectral cutoff of blazars obtained by the LAT will be used to measure the evolution of the extragalactic background light [28]. This will provide vital information complementary to that by ground Cherenkov telescope (e.g., [29]) on star formation history in early universe

• Fig. 6.

Simulation of a GLAST

• bservation of spectral evolu-

tion of 3C 279 flare in 1996, taken from [15]. A model of light curve consistent with the EGRET data is assumed and shown as blue line. Red and green data points are the mea- sured photon index and the flux, respectively.

References [1] G. Kanbach et al. 1988, Space Sci. Rev. 49, 69 [2] D. J. Thmpson et al. 1993, ApJS 86, 629 [3] P. Michelson et al. 2007, in preparation [4] W. B. Atwood et al. 2000, NIMA 446, 444 [5] do Coute e Silva et al. 2001, NIMA 474, 19 [6] D. J. Thompson et al. 2002, IEEE. Trans. Nucl. Sci. 49, 1898 [7] T. Mizuno et al. 2004, ApJ 614, 780 [8] T. Ohsugi et al., NIMA 541, 29 [9] W. Atwood et al. 2007, in preparation [10] J. Ormes et al. 2007, P17.11, poster in GLAST Science Symposium at Stan- ford University

9

[11] A. Moiseev et al. 2007, P17.15, poster in GLAST Science Symposium at Stanford University [12] E. Grove et al. 2007, in preparation [13] A. Moiseev et al. 2007, Astroparticle Physics 27, 339 [14] G. G. Lichti et al., AIP conference proceedings 906, 119 [15] GLAST Science Brochure (http://glast.gsfc.nasa.gov/science/). [16] de Jager et al. 1996, ApJ 457, 253 [17] F. Aharonian et al. 2006, A&A 460, 365 [18] J. K. Daugherty et al. 1996, ApJ 458, 278 [19] R. Romani et al. 1996, ApJ 470, 469 [20] M. Razzano et al. 2007, P14.17, poster in GLAST Science Symposium at Stanford University [21] K. Koyama et al. 1995, Nature 378, 255 [22] F. Aharonian et al. 2006, A&A 464, 253 [23] F. Aharonian et al. 2007, ApJ 661, 236 [24] S. Funk et al. 2007, P14.6, poster in GLAST Science Symposium at Stanford University [25] R. C. Hartman et al. 1999, ApJS 123, 79 [26] D. Donato et al. 2001, A&A 375, 739 [27] A. E. Wehrle 1998, ApJ 497, 178 [28] A. Chen et al. 2004, ApJ 608, 686 [29] F. Aharonian et al. 2006, Nature 440, 1018