iact swiss participation

8/19/2019 Iact Swiss Participation

1/20

The Cherenkov Telescope Array Swiss participation and perspectives


2/20

THE CHERENKOV TELESCOPE ARRAY SWISS PARTICIPATION AND PERSPECTIVES

Executive Summary

CTA, the Cherenkov Telescope Array, is the most ambitious next

generation experiment in ground based gamma-ray astrophysics. As the

successor of the H.E.S.S. and MAGIC instruments, it brings together the

European particle physicists and high energy astrophysicists communities.

CTA is highly ranked in the ASPERA and ASTRONET roadmaps and is

listed as an emerging project in the ESFRI roadmap of 2006.

The CTA collaboration – more than 30 research institutes – is now

organizing itself. This document presents a plan for the Swiss contribution

to CTA and perspectives for the future as defined by the four leading

institutes in the field in Switzerland: the INTEGRAL Science Data Centre

and Department of Nuclear and Particle Physics at the University of

Geneva, the Institute for Particle Physics at ETH Zurich and the Paul

Scherrer Institute.

Swiss scientists have the ambition to lead the effort behind the

establishment of the Cherenkov Science Data Centre and the construction

of a novel solid-state camera prototype for CTA.


3/20


The segmented mirror of the H.E.S.S. Cherenkov telescope

CONTENTS......................................................................A breakthrough in TeV astrophysics# 2

...........................................................................The Cherenkov Telescope Array# 4

.........................................................................................Astrophysics with CTA# 5

! ..........................................................................-ray astrophysics in Switzerland 7

..................................................................................Swiss participation in CTA# 11

...................................................................................................Cost estimates# 15

...............................................................................Perspectives and impressum# 16

Page 1


4/20


In March 2007, the High Energy Stereoscopic System(H.E.S.S.) project was awarded the Descartes ResearchPrize of the European Commission for offering “A newglimpse at the highest energy Universe”. Together withthe instruments MAGIC and VERITAS (in the northernhemisphere) and CANGAROO (in the southernhemisphere, like H.E.S.S.), a new wavelength domainhas opened for astronomy, the domain of very highenergy gamma-rays (0.1–100 Tera electron volts; 1011–1014 eV), energies which are a million million timeshigher than the energy of visible light.

Radiation at these energies differs fundamentally fromthat detected by astronomical instruments at lowerenergies: GeV to TeV gamma-rays cannot conceivablybe generated thermally by emission from hot celestialobjects. The energy of thermal radiation reflects thetemperature of the emitting body, and apart from theBig Bang there is nothing hot enough to emit such

gamma-rays in the known Universe. Instead, highenergy gamma-rays probe a “non-thermal” Universe,where mechanisms other than thermal emission by hotbodies allow the concentration of large amounts ofenergy into a single quantum of radiation. High energygamma-rays can be produced in a top-down fashion bydecays of heavy particles such as the hypothetical darkmatter particles or cosmic strings, both relics whichmight be left over from the Big Bang. In a bottom-upfashion, gamma-rays can be generated when highenergy nuclei – accelerated for example in giganticshock waves created in stellar explosions – collide with

ambient gas particles. The flux and energy spectrum ofthe gamma-rays reflect the flux and spectrum of the highenergy nuclei. They can therefore be used to trace thesecosmic rays in distant regions of our own Galaxy oreven in other galaxies.

Imaging the Universe

The first images of the Milky Way in very high energygamma-rays were obtained in recent years, and reveala chain of gamma-ray emitters lining the Galactic plane,

demonstrating that sources of high energy radiation areubiquitous in our Galaxy. Sources of this radiationinclude supernova shock waves, where atomic nucleiare presumably accelerated and generate the observedgamma-rays. Another important class of objects in thiscontext are nebulae surrounding pulsars, where strongrotating magnetic fields give rise to a steady flow ofhigh energy particles.

Some of the objects disco-vered are binary systems,where a black hole or neutron

star orbits a massive star.Along the elliptical orbit, thecond i t i on s f o r pa r t i c l eacceleration vary and hencethe radiation intensity ismodulated with the orbitalperiod. These systems arepa r t i cu la r l y i n t e r e s t i ngbecause they allow the studyof how particle accelerationprocesses respond to the

varying ambient conditions.One of several surprises was

the discovery of “dark sources”, objects which emit highenergy radiation, but have no obvious counterpart inother wavelength regimes. In other words, there areobjects in the Galaxy which are so far only visible anddetectable in high energy gamma-rays. Beyond our Galaxy, well over a dozen extragalacticsources of high energy radiation have been discovered,located in active galaxies, where a super massive blackhole at the centre is fed by a steady stream of gas andis releasing enormous amounts of energy. Gamma-raysare believed to be emitted from the vicinity of these

A breakthrough in TeV astrophysics

Page 2

The Milky Way viewed in very high energy gamma-rays (bottom panel), compared to itsappearance in visible and infrared light


5/20


black holes, allowing the study of the processesoccurring in this violent and as yet poorly understoodenvironment.

Cherenkov technique

The recent breakthroughs in very high energy gamma-ray astronomy were achieved with Cherenkovtelescopes. When a gamma-ray enters the atmosphere,it interacts with atmospheric nuclei and generates acascade of secondary particles through the atmosphereat speeds higher than the speed of light in the gas.These particles emit a beamed bluish light, theCherenkov light. The Cherenkov photons are emitted ina pulse which lasts about 10 ns. They illuminate an areaon the ground of about 250 m in diameter. Opticaltelescopes can be used to image the particle cascade

by detecting the secondary Cherenkov emission andhence reconstruct the trajectory of the initial gamma-ray.Large optical reflectors with areas in the 100 m2 rangeand beyond operated at dark sites are required tocollect enough Cherenkov photons to image the particlecascades. An image of the gamma-ray sky can becreated by accumulating trajectory data from manyphoton induced cascades.

The imaging atmospheric Cherenkov technique waspioneered by the Whipple Telescope in the UnitedStates. After more than 20 years of development, thefirst source of very high energy gamma-rays, the CrabNebula, was detected in 1989. The Crab Nebula is

among the strongest sour-ces of very high energygamma-rays, and is oftenused as a standard candle.Modern instruments, usingmultiple telescopes to trackthe cascade from differentperspectives and emplo-ying fined-grained photondetectors for improvedimag ing , can de tec t

sources down to 1% of theflux of the Crab Nebula.The use of “stereoscopic”telescopes to provideimages of the cascadefrom different viewing

points was pioneered by the European HEGRAinstrument, precursor of the current H.E.S.S. andMAGIC telescopes.

Page 3

TeV images of shock fronts in supernova, revealing cosmic acceleration of energetic particles

Alignment of the mirrors of the MAGIC telescope


6/20


The Cherenkov Telescope Array

Page 4

A new research infrastructure

In the field of very high energy gamma-ray astronomy,Europe with H.E.S.S. and MAGIC holds a clear leadingposition. The spectacular astrophysics results fromcurrent Cherenkov instruments have generatedconsiderable interest in both the astrophysics andparticle physics communities and have spawned theurgent wish for a next-generation, more sensitive andmore flexible facility, able to serve a large community ofusers. The proposed CTA facility – an array of

Cherenkov telescopes building on proven technologydeployed on an unprecedented scale – will allow theEuropean scientific community to remain at the forefrontof the advancement of research. CTA is a new facility,well beyond the conceivable upgrades of existinginstruments such as H.E.S.S. or MAGIC. The CTA projectfor the first time unifies the research groups working inthis field in Europe in a common strategy, resulting in aunique convergence of efforts, human resources, andknow-how. Interest in and support of the project spanscientists from more than 30 research institutes all overEurope, in the Czech Republic, Finland, France,Germany, Italy, Ireland, the Netherlands, Poland,Spain, Switzerland, and the United Kingdom, wishing touse such a facility for their research and willing tocontribute to its design and construction. CTA will offer

worldwide unique research services to users fromdifferent countries. The number of young scientistsattracted to the still young field of gamma-rayastronomy is growing at a steady rate, drawing fromother fields such as nuclear and particle physics, inaddition to increased interest by other parts of theastrophysical community, such as radio and X-rayastronomers. For the first time in this field, CTA willgenerate large amounts of data open to public access,allowing data mining in addition to targetedobservation proposals. Accessible through policies and

tools in use in the astronomical observatories, CTA aimsto emerge as a cornerstone in a networked multi-wavelength, multi-messenger exploration of theUniverse.

CTA Performance

Imaging atmospheric Cherenkov radiation is by now awell established technique and, at least in the coreenergy range above 100 GeV, the performance andthe limitations of current instruments are wellunderstood. This allows a reliable extrapolation towardsa next-generation instrument, providing vastly improvedperformance and increased flexibility to accommodatea large community of users. The CTA observatory aimsat increasing sensitivity in the core energy range fromabout 100 GeV to about 10 TeV by roughly one orderof magnitude, and at the same time aims at expandingthe energy range for very high energy gamma-rayastronomy towards both lower and higher energies,effectively increasing the usable energy coverage by afactor of ten. Furthermore CTA will provide bothsignificant improvements in angular resolution, revealing

finer details in the sources, and unprecedenteddetection rates, enabling researchers to track transientphenomena on very short time scales.

.

.

.

.

’ . . .

.

.

. .

.

. .

.

:

, , .

, , .

Possible telescope configuration for CTA (top) together withother observatories of the 2020s: SKA, ALMA, ELT and XEUS

CTA sensitivity (50h)compared to existinginstruments


7/20


The high energy phenomena, which can be studied withCTA, span a wide field of galactic and extragalacticastrophysics, of plasma physics, particle physics, astro-particle physics, and fundamental physics of space-time.They encode information on the birth and death ofstars, on the matter circulation in the Galaxy, and onthe history of the Universe.

Supernovae remnants, pulsar wind nebulae, and cosmicrays. A paradigm of high energy astrophysics is thatcosmic-rays are accelerated in supernova explosion

shocks. While particle acceleration up to energies wellbeyond 1013 eV has now clearly been demonstrated byH.E.S.S., it is by no means proven that supernovaeaccelerate the bulk of cosmic rays. The large sample ofsupernovae which will be observable with CTA, in somescenarios several hundred objects, and in particular theincreased energy coverage towards lower and higherenergies will allow sensitive tests of acceleration modelsand determination of their parameters. Pulsar windnebulae surrounding the pulsars created in supernovaexplosions are another abundant source of high energyparticles, including potentially high energy nuclei.Energy conversion within pulsar winds and theinteraction of the wind with the ambient medium and thesurrounding supernova shell challenge current ideas inplasma physics.

Pulsar physics. Pulsar magnetospheres are known to actas efficient cosmic accelerators, yet there is no acceptedmodel for this particle acceleration, a process whichinvolves electrodynamics with very high magnetic fieldsas well as the effects of general relativity. Pulsedgamma-ray emission allows the separation of processes

occurring in the magnetosphere from the emission of thesurrounding nebula. Competing models predictcharacteristic cut-off features in the spectra of pulsedgamma-rays in the low-energy range of CTA. Comparedto gamma-ray space instruments, CTA with its muchlarger detection rate is not affected by glitches in pulsarperiods which may compromise periodicity measure-ments requiring very long integration times.

Micro-quasars and X-ray binaries. Three very highenergy gamma-ray emitters are currently known whichare binary systems, consisting of a compact object – aneutron star or black hole – orbiting a massive star.Whilst many questions are open concerning gamma-rayemission from such systems – in some cases it is not

even clear if a pulsar-driven nebula around a neutronstar or accretion onto a black hole is the energy source– it is evident that they offer a unique chance to“experiment” with cosmic accelerators. Along theeccentric orbits of the compact objects, the environment(including crucially the radiation field) changesperiodically, resulting in a modulation of the gamma-rayflux, allowing the study of how particle accelerationreacts to these environmental conditions. Equallyinteresting, the physics of micro-quasars in our ownGalaxy resembles the processes occurring aroundsupermassive black holes in distant active galaxies,except for the much faster time scales, providing insightsinto these mechanisms.

Stellar clusters and stellar systems. While the classicalparadigm emphasizes supernova explosions as thedominant source of cosmic rays, it has been speculatedthat cosmic rays arealso accelerated instellar winds aroundmassive young stars

before they explodeas supernovae, oraround star clusters.Indeed , t here i sgrowing evidence inexisting gamma-raydata for a populationof sources related toyoung stellar clustersand environmentswith strong stellarwinds. However, lack of instrument sensitivity currentlyprevents the detailed study and clear identification ofthese sources of gamma radiation.

Page 5

Astrophysics with CTA

Eta Carinae generates shocks whichaccelerate particles

Variable TeV signal from the X-ray binary LSI +61 303


8/20


The Galactic Centre. The Galactic Centre hosts thenearest supermassive black hole, as well as a variety ofother objects likely to generate high energy radiation,including hypothetical dark matter particles which maypair-annihilate and produce gamma-rays. Indeed, the

Galactic Centre hasbeen detected as asource of high energygamma - rays , andindications for highenergy particles dif-

fusing away from thecentral source andilluminating the densegas clouds in thecentral region havebeen detected. Inobservations with impr-oved sensitivity andresolution, the Galac-tic Centre provides arich reservoir of inter-

esting physics from particle acceleration via the not wellknown diffusive propagation of cosmic-ray particles toexotic phenomena such as acceleration and curvatureradiation of protons at the edge of rapidly spinningblack holes.

Active galaxies. The supermassive black holes at thecores of active galaxies are known to produce outflowswhich are strong sources of high energy gamma-rays.The fast variability of the gamma-ray flux on minute timescales indicates that gamma-ray production must occurnear the black hole, assisted by highly relativistic motion

resulting in a contraction of time scales when viewedfrom an observer on Earth. Details of how these jets arelaunched by the black hole and even the kinds ofparticles of which they consist are poorly understood.Multi-wavelength observations with high temporalresolution can distinguish different scenarios, but are atthe edge of the capability of current instruments.

Galaxy clusters. Galaxy clusters act as storehouses ofcosmic rays, since all cosmic rays produced in clustergalaxies since the beginning of the Universe will beconfined to the cluster. Probing the density of cosmicrays in clusters via their gamma-ray emission thusprovides a calorimetric measure of the total integratednon-thermal energy output of galaxies. Accretion or

merger shocks in clusters of galaxies provide anadditional source of high energy particles. Emissionfrom galaxy clusters is predicted at levels just below thesensitivity of current instruments.

Cosmic radiation fields and cosmology. Via theirinteraction with extragalactic light, high energy gamma-rays from distant galaxies allow the extraction ofcosmological information on the density of light inextragalactic space and therefore about the formationhistory of stars in the Universe. Gamma-rays experience

an energy-dependent attenuation when propagatingthrough intergalactic space, due to electron-positronpair production on the extragalactic background light.This phenomenon allows determination of extragalacticlight levels, unimpeded by the overwhelming amount offoreground light from the solar system and the Galaxy,which makes direct measurements prone to very largesystematic uncertainties. Pair-production halossurrounding active galaxies may even allow measu-rement of the evolution of light intensity with redshift.

Search for Dark Matter . The dominant form of matter inthe Universe is a yet unknown type of dark matter, mostlikely in form of a new class of particles such aspredicted in supersymmetric extensions to the standardmodel of particle physics. Dark matter particlesaccumulate in, and cause, wells in gravitationalpotential, and with high enough density they arepredicted to have annihilation rates resulting indetectable fluxes of high energy gamma-rays. CTAwould provide a sensitive probe of this annihilationradiation, and will help to verify if such particles –which by then might be discovered at the Large Hadron

Collider LHC – form the dark matter in the Universe.

Probing space-time. Due to their extremely shortwavelength and long propagation distances, very highenergy gamma-rays are sensitive to the microscopicstructure of space-time. Small-scale perturbations of thesmooth space-time continuum, as predicted in theories ofquantum gravity, should manifest themselves in a (tiny)energy dependence of gamma-ray propagation speeds.Burst-like events of gamma-ray production, e.g. in activegalaxies, allow this energy dependent dispersion ofgamma-rays to be probed and can be used to placelimits on certain classes of quantum gravity scenario,and may possibly lead to the discovery of effectsassociated with quantum gravity.

Page 6

TeV sources in the galactic centre(top) and diffuse emission (bottom)


9/20


The high energy group of the astronomicalobservatory of the University of Geneva started itsactivities in 1988. The group was selected by theEuropean Space Agency in 1995 to build and operatethe INTEGRAL Science Data Centre. The group thenmoved from the Geneva observatory to a beautifulclose-by setting in Ecogia.

INTEGRAL and its science data centre

While designing the gamma-ray observatory of theEuropean Space Agency, INTEGRAL, it was decidedthat the science data centre should be provided by thescientific community in order to best match the scienceanalysis functionalities with the needs of the scientists.This allowed the community to use the competence andthe willingness of institutes and funding authorities toactively contribute to this important element of themission. This decision lead to the creation of ISDC, a

centre to which a consortium of 12 institutes in Europeand the United States have contributed. The mainfunding source is provided by the Swiss authorities andthe State of Geneva and its University.

Since the launch of INTEGRAL (October 17, 2002), thedata are transmitted continuously from the detectorson-board to ISDC. The data are analyzed three timesat ISDC, within seconds to detect gamma-ray bursts,within hours to detect new sources in the sky orparticular variations of sources and within weeks toextract the most information possible from the data

and distribute it to the science community world-wide.Following the success of the INTEGRAL operations,ESA selected ISDC and Geneva observatory toperform part of the processing for the Planck and Gaiamissions, to be launched in the next years. ISDC hasalready been chosen by the scientific community toprovide the science data centre for the next majoreuropean X-ray observatory, XEUS.

Some 40 people from more than 10 countries worktogether at the ISDC. The team is made of some 15-20

engineers and technicians, 14 PhD scientists, 4-5 PhDstudents and 3 administrative staff. They developed

and now operate a large range of complex softwarethat exceeds in their capabilities all the originalexpectations. ISDC has also benefited from manycontributions from outside engineers and scientists, inparticular those associated with the teams that built theINTEGRAL instruments.

High energy astrophysics at the ISDC

In parallel to their various operational duties, scientistsare active in different research fields ranging from thestudy of various types of X-ray binaries, pulsars, stellarformation, massive stars, supernova remnant, activegalactic nuclei, galaxy clusters, gamma-ray bursts anddark matter/cosmology using all types of astronomicalfacilities. Since 1998, 11 PhD thesis have beenconducted with 4 additional ones on-going. During thelast years, ISDC scientists contributed in 50 referredpapers per year on average. ISDC scientists

participate in the CTA, XEUS and POLAR projects.

!-ray astrophysics in Switzerland

INTEGRAL Science Data Centre (ISDC) – University of Geneva

The INTEGRAL spacecraft

Page 7


10/20


For more than ten years, the experimental researchactivities at the Institute for Particle Physics concentrateon two main pillars: particle physics at accelerators andastro-particle physics.

Research groups at IPP have a long-term tradition ofsuccessful participation in experiments at the highenergy frontier, from the design to prototyping,

construction, operation and data analysis of large-scaleexperiments like the L3 and LHC experiments at CERN.

IPP is contributing in a major way to the CMS experi-ment at LHC.

Astro-particle physics at IPP

IPP members initiated the L3+Cosmics experiment atCERN. From the data collected until November 2000,a variety of important results have emerged, related to

cosmic rays and astro-particle physics. IPP also playeda leading role in the design and construction of theAMS detector. A prototype (AMS-01) was installed onthe space shuttle Discovery during a 10 days mission in1998 and demonstrated the feasibility to operate alarge magnetic spectrometer in space. AMS-02 shouldbe installed at the International Space Station in thenear future. One of the main objectives is to investigatewhether antimatter exists in the Universe today in ameasurable quantity.

Since 2003 IPP participates in MAGIC, a Cherenkov

telescope experiment observing very high energygamma-rays from galactic objects such as supernovaremnants, pulsars or binary systems, as well as extra-galactic objects, most notably Active Galactic Nucleiand Gamma-Ray Bursts. IPP has been involved in thecommissioning of the MAGIC-I telescope, theimprovement of the Active Mirror Control (AMC) andstrongly contributes to data analysis. In addition, IPP isresponsible for the design and construction of the AMCfor MAGIC-II, presently under construction on theCanary Island of La Palma.

Research and development performed in collaborationwith PSI on novel photo detectors (gAPDs) demonstra-ted the feasibility to use this new device for asubstantially improved camera to detect the Cherenkovlight from high energy gamma-rays. The next step is tobuild a prototype camera using gAPDs and newreadout electronics in view of a possible futureapplication for CTA.

IPP is also strongly involved in the CTA studies since thevery beginning. We are represented in the adhoc CTAsteering committee and participate in several CTAworking groups.

Institute for Particle Physics (IPP) – ETH Zurich

The MAGIC telescope located at the Roque de los MuchachosObservatory on the Canary Island of La Palma

Page 8


11/20


The Department of Nuclear and Particle Physics of theUniversity of Geneva has a track record in excess of 15years concerning the design, manufacture andexploitation of silicon tracking devices, calorimetersand scintillation detectors for particle physics. Recentprojects include the central tracking chamber and thesilicon vertex detector for the L3 experiment at CERN,the silicon detector for the NASA AMS and the CERN

ATLAS experiments as well as the complete hardwarefor the FAST experiment at PSI.

Astro-particle physics at DPNC

DPNC participated to the LEP experiment L3 whichmeasured the Zo resonance and the number of familiesof leptons existing in nature as well as limits on themass of the Higgs boson. These results have profoundimplications on the big bang theory and on cosmology.

DPNC is a founding member of the collaborationbuilding and operating the Alpha MagneticSpectrometer (AMS), which will analyze cosmicradiation of galactic and extragalactic origin at analtitude of about 400 km above Earth. A precursorflight on NASA Space Shuttle mission STS-91 alreadytook place in June 1998, using the prototype detectorAMS-01.

Based on this experience, a new detector wasconceived which is being completed. The detector isscheduled to be 'launch ready' at the end of the year

2008. The University of Geneva group hasresponsibilities for the assembly and integration of themain detector, i.e. the Silicon tracking detector, and forits associated read-out electronics.

DPNC is also collaborating with the ISDC in theframework of the POLAR project, a Compton camerameasuring the polarization of photons from gamma-raybursts. The scintillating bar technology for this device isbased on the hardware of the Fibre Active ScintillatorTarget (FAST) experiment at the Paul Scherrer Institute

which aims at measuring the muon lifetime and the

strength of the weak interaction with a precision of10-6.

The DPNC astro-particle physics group has alsocontributed significantly to the analysis of data fromthe AMS-01 mission and the preparation of AMS-02data analysis. The group has specialized on thesubjects of heavy ion detection and photon detection

by conversion in the AMS spectrometer.

DPNC – University of Geneva

Page 9

Insertion of the AMS tracking detector into the magnetstructure

FAST experiment at the Paul Scherrer Institute


12/20


The Paul Scherrer Institute (PSI) is one of the leadingEuropean laboratories pushing forward thedevelopment of novel particle detectors and associatedreadout electronics for high energy physics, which areused in other fields of research as well. PSI has madeimportant hardware contributions to the spaceastronomical missions XMM-Newton, RHESSI, the

James Webb Space Telescope and POLAR.

PSI has developed two technologies which are keyelements for a novel CTA camera.

Solid state photo detectors

The photo sensors in the electromagnetic calorimeter ofthe CMS experiment at LHC have to operate in anextremely harsh environment. CMS with PSI as leadinginstitution developed in close collaboration withHamamatsu Photonics a radiation hard avalanche

photodiode (APD). Some 100 prototypes have beencharacterized and during the mass production of thefinal device 140 thousand APDs have been subject to ascreening procedure in order to ensure a reliability of99.9%.

The recently developed Geiger-mode avalanchephotodiodes (gAPDs) have a substantially improvedsensitivity and can detect single photons with a highefficiency of more than 60%. They have propertiessimilar or even superior to photomultiplier tubes andare therefore also named silicon photomultiplier.

Advantages of this novel device are a high internalgain (105 to 106), the small dimension, a relative lowbias voltage (


13/20


The high energy astrophysicists at ISDC and the particlephysicists at IPP, DPNC and PSI plan to have a strongSwiss participation in CTA. ISDC, together with IPP andDPNC, plan to lead the effort behind the establishmentof the science data centre for CTA. In addition, with theexpertise of PSI, these institutes plan to lead theconstruction and scientific exploitation of a novel solid-state camera prototype for CTA.

Cherenkov Science Data Centre

Switzerland is in a unique position to host theCherenkov Science Data Centre as the ISDC and IPPteams together combine the knowledge of theCherenkov telescope techniques, construction andcalibration and of the organization, build-up andoperation of a worldwide data centre active in highenergy astrophysics. The team also represents the needsand interests of both the astronomical and astro-particlecommunities.

CTA will provide a high quality set of datacomplimentary to those obtained with other majorastronomical facilities on the ground and in space.Many problems in modern astronomy require the use ofseveral of these facilities together in order tounderstand the physics of the objects under study.Within the next 10–15 years, Switzerland will host acentre which will make data from major European highenergy astronomical observatories (CTA, INTEGRAL,XEUS) available in inter-operable format and providean integrated level of service to the scientific community.This centre will foster scientific research in high energy

astrophysics in the world and in Switzerland and willenhance the world-wide visibility obtained bySwitzerland through ISDC.

CTA, the first TeV observatory

In the 1980’s, after 20 years of developments, X-rayastronomy moved from PI led experiments toobservatories open to the scientific community at large.This evolution led to the enormous success of X-rayastronomy which is presently one of the major branchof astronomy. Ten years ago the hard X-ray – MeVcommunity followed the same route and, now, it is theturn of the GeV–TeV community, as illustrated by the

extraordinary interest triggered by the recent results ofH.E.S.S. and MAGIC.

CTA will be the first Cherenkov telescope conceivedfrom the start as an observatory. The CTA projectmerges various teams organized around PI ledexperiments with a larger community to build the mostambitious world-class observing facility in the field.

The CTA observatory will have most of its observingtime available as a guest observer program open to the

scientific community at large. Principal investigatorsfrom any country worldwide can submit observingproposals in response to announcements of oppor-tunity. Proposals can be submitted to request normalobservations, observations coordinated with otherobservatories, target of opportunity and will beselected based on scientific merit through a peer-reviewprocess.

The data resulting from an observation will be madeavailable to the guest observer during a proprietaryperiod before becoming public. The data, calibrationand software will be open and easily manageable toallow any PhD student to analyze CTA data and publishresults. We estimate that CTA will detect a thousand

Swiss participation in CTA

Page 11

Multi-mission source archive prototype at the ISDC


14/20


sources and will conduct of the order of a hundredobservations per year. The community of CTA users(observers and archive users) will be a few hundredscientists. These numbers are similar to what ISDC isused to manage for INTEGRAL, in fact the CTA andINTEGRAL user communities overlap to some extend.

Policies, time allocation and science operations

One of the roles of the science data centre, togetherwith the observatory governing bodies, is to establish

policies, documentation and software tools which willallow potential observers to access the observatory,first by submitting proposals. Additional tools areneeded to manage the proposals, organize andsupport a time allocation committee, manage andschedule observations and support the community forthe organization of target of opportunity observationsand multi-wavelength campaigns. Finally, monitoring ofthe observatory usage and performance and reportingis an essential aspect to support the observatorygoverning and financing bodies. ISDC scientists have abroad view of all these aspects by participating closelyto the INTEGRAL operations and acting as proposers ortime allocation committee members for a wide selectionof observatories.

On-site data centre activities The sky at GeV-TeV energies is much more variablethan in visible wavelengths. Some sources shine duringlimited periods of time (minutes-hours) and must bestudied during these periods to organize follow-upobservations.

The CTA telescopes will generate about 10 TeraBytes ofraw data per observing night and will be located onone or two remote sites. These sites will most probably

lack broad-band network connections which would benecessary to transfer even reduced data (10 GB/night)in near real time to the data centre for analysis. As aconsequence, a quick-look data analysis will have to beperformed on-site in near real-time. This analysis will beused to check the quality of the data and detectwhether major events have taken place in the patch ofthe sky observed by CTA. In case of detection of newsources or of unexpected events, an alert will begenerated which will help, where appropriate, tomodify the foreseen observation plan or trigger follow-

up observations with other observing facilities. Someadjustments may also be needed according to theweather reports. Some level of interaction with theobserver during the data taking should also be plannedand in case of special scientific needs data products willneed to be distributed to the observer very quickly. Foran observatory of the sensitivity of CTA, with thepossibility that various parts of the telescope arraycould point to various targets, one of the on-sitescientists is expected to work full-time for the activitieslisted above. Each week a package of about 100 tapeswill be shipped to the science data centre. A temporaryarchive corresponding to two months of data will beneeded on-site.

Raw data archive

The raw data tapes received in Europe will be checkedand a backup will be created. Finally the data will beuploaded to the raw data archive. The growth rate ofthis archive is estimated to be of the order of threePetaBytes per year. Hence, the raw data storagerepresents the most significant hardware cost in the CTA

data processing chain. The most cost effective solutionwill probably be based on tape robots and a diskstaging area. This will need to be revisited beforeimplementation.

A standard data reduction chain for the raw data willbe developed. The total size of the output data is afactor 1000 smaller than the raw data size. This rendersthe storage and handling of the input data for thehigher level processing relatively simple even at today'sstandards. A mere volume of three TeraBytes per yearwill have to be stored and handled during theoperational phase. Some physics goals however, willonly be achievable by directly accessing the raw dataand do require a lot of expert knowledge. This together

Page 12

Instrument response &calibration

Air shower simulations

End User Analysis

Raw Data Analysis

Raw Data:triggered events

Parametrized events

Photon cand. events

High level data:images, spectra,

lightcurves

CTA data processing(simplified)


15/20


with the standard raw data processing implies that theraw data will be read by a small and well identifiednumber of users. Apart from the high volume of inputdata, the amount of CPU power required for the rawdata processing is moderate. We estimate that a smallcluster with some 10 nodes will be sufficient.

Calibration, standard processing and science archive The community of CTA users will not be limited to thosemost familiar with the instrumentation, but will span

astronomers of different fields. This imposes that thedata be made accessible in a format which can easilybe understood and with the support necessary to allowtheir full scientific exploitation. A further importantaspect of CTA is that its data will be used in differentresearch projects during many years. It is, therefore,necessary to build an archive of the CTA products. Thisarchive will include event lists in a format that is easilytransportable as well as the results of a standard dataanalysis.

A data model will be developed together with astandard analysis pipeline which will use the raw datato produce calibrated gamma-ray events together witha set of standard scientific results for each observation.The data model and the software can be built using thesame concepts and basic tools developed by ISDC forINTEGRAL and other missions, reducing the overallimplementation and maintenance costs. All the dataproducts will be archived and made available first toguest observers and then to the public. Public dataconcerning specific sources in the sky will be madeavailable with data from INTEGRAL, XEUS and other

experiments. This source archive concept is alreadybeing prepared for INTEGRAL at ISDC.

Large observatories and scientific instruments developand last for much longer than usual software life time.The analysis software needs therefore to be open andvery simple and shall avoid any short lifetime productsdeveloped in the community or by the industry. Thedata products can of course be interfaced with the mostpopular fancy interfaces, tools or concepts (Virtualobservatories, GRID, specific analysis packages) for arelatively small effort. The primary responsibility of thescience data centre is to provide well calibrated andusable data and dedicated software tools and not todevelop short-lived software technologies.

Off-line analysis and user support

The data centre has to provide to the CTA users a copyof its analysis software which will allow scientists torepeat, customize and refine the standard analysis andextract the relevant information from the data forpublication. The open data format will also allowscientists to use their own or other tools to reduce thedata as well as to use the standard tools developedover the years in the scientific community for imaging,spectral and time series analysis.

The analysis software will be completely independentfrom any infrastructure, to insure that any componentcan be used everywhere without the data centresystem. The distributed software will of course beavailable free of any license fee and ported on themajor user platforms. This is a standard practice, inparticular enforced for INTEGRAL at ISDC. The input tothe off-line analysis tools are the standard dataproducts retrieved from the CTA archive. The user shallbe able to handle standard data of different levels(from raw data to the high-level data products). Takinginto account that different observers will have differentlevel of expertise in CTA data analysis, the users shouldhave the possibility to call most off-line analysis toolsfrom a comprehensible graphical user interface, inwhich all or only essential parameters are requested,with other parameters set to best-guess default values.

The data centre will also need to support the CTA usersand help new users to use the facility by providing ahelp-desk, yearly data analysis training courses and anewsletter.

Data from other Cherenkov arrays

The data obtained by the past and current Cherenkovarrays (in particular MAGIC and H.E.S.S.) will need tobe made accessible to the scientific community in thesame format as and interoperable with the CTA data.These data will represent only a small fraction of theCTA data volume, however some specific work will needto be performed to make them freely accessible in anopen data format. This work is essential to build anhistoric archive of relevant TeV data that could be usedfor future scientific research. The work on the data ofHEGRA is already on-going at ISDC.

Page 13


16/20


Solid-State Cherenkov Camera

The heart of a Cherenkov telescope is the light detectionsystem, a camera for very low intensities, and theassociated electronics. It typically consists of Winstoncone light concentrators, photomultiplier tubes andanalogue electronics to condition the signals. A crucialparameter of the light detection system is the quantumefficiency of the photo detectors. For photomultipliertubes, the probability to detect an incident photon

hitting the photocathode varies between 25 and 35%for the wavelength range of the Cherenkov light. DPNCof University of Geneva has extended usage andintegration experience with the latest metal mesh photo-multipliers from the FAST project.

An important step forward in photo-detection isrepresented by the recent development of Geiger modeAvalanche Photo-Diodes (gAPD) or silicon photo-multiplier (SiPM). gAPDs promise photon detectionefficiencies of more than 60%, a factor two higher thanphotomultiplier tubes. The Paul Scherrer Institute is one

of the leading European laboratories pushing forwardthe development of APDs for usage in high energyphysics. In the framework of the CMS electromagneticcalorimeter project, 140 thousand APDs have beencharacterized, integrated, calibrated and commissionedinto an LHC detector. PSI continues to work in closecontact with industry to help develop and improvegAPDs for optical applications in physics.

In the context of CTA, we propose to develop a cameraprototype based on gAPD technology. To be realistic,the prototype must contain of the order of 1000 pixels,consisting of several gAPDs each, with completeinfrastructure, mechanics and adapted readout

electronics. Qualification as a candidate technology formassive use in CTA also requires usage of the cameraunder realistic conditions. A simple possibility is to

integrate the prototype into a refurbishment of theHEGRA CT3 telescope on the La Palma site. Anotherpossibility is to install it on one of the telescopes whichwill be developed as prototype for CTA. Our camerawill therefore serve not only for technology testingpurposes, but will allow scientific observations.

At this point in time, gAPD chips are produced with lowyield and thus at rather high end-user cost. Part of theproject will thus be devoted to work with industry on theimprovement of the manufacturing procedures towards

lower unit cost. However, the prototype camera will usecurrently available chips, to study in parallel theoptimum integration into the camera and to developreadout electronics based on the DOMINO chip designfrom PSI. Integration studies will include an assessmentof temperature and high voltage stabilization,mechanical tolerances and protection against impactsfrom the environment.

A central issue will be to establish a modular approachthat can be readily industrialized for mass production.For this purpose, we propose to work closely with the

Swiss industry from the beginning. The aim is to be in acompetitive situation when the decision on cameratechnology for the CTA telescopes will be taken.

The table below shows the estimated cost of the gAPDcamera prototype development, totaling about 1.3 M$.The total development should extend over no longerthan two years and end up with a long-term test underrealistic conditions inside the refurbished CT3 telescopeor an equivalent mount.

Page 14

Manpower[FTE]

Hardware[k$]

gAPD detectors

Winston cones

Front-end electronic

Trigger logic

Housing

Power supply

Total

1.0 300

50

0.7 120

0.7 150

1.0 80

1.0 150

4.4 850

Cost estimate for the prototype gAPD camera

APD detectors developedfor CMS at PSI, observedthrough a scintillator bar


17/20


The cost of the various data centre activities have beenevaluated using the experience of ISDC in building andrunning the data centre for INTEGRAL; the experienceof the H.E.S.S. and MAGIC teams; the experience ofESA and ESO, running time allocation committees,observation scheduling and data transfers; and theexperience of CERN in managing Peta-Bytes archives.The costs have been estimated taking into account thatthe data centre is developed in coordination with thoseof other missions, in particular INTEGRAL and XEUS,which offers cost savings, optimized long term

maintenance and operations, solidity and stability. Notethat scientific research by PHD students and youngpostdocs is not included.

The yearly costs are presented in the table below forvarious data centre activities in terms of manpower(FTE) and hardware costs and differentiated for thedevelopment and operational phases. They assume adevelopment of 3 years before operations and a singlesite for CTA. These figures do not include the localoffice space and running costs (to be made available bythe University of Geneva) nor the on-site housing.

During the cost analysis each activity was split invarious tasks. At that stage the cost of each of the taskscould have an uncertainty as large as 30%. Theaccuracy of the total cost, summing all activities,reaches 10%. The hardware costs are dominated by theraw data archive whose size (and cost) will depend onthe detailed design of CTA. The spending profile forboth the Cherenkov science data centre and the

prototype gAPD camera are presented in the figureabove and amount to about 2.5M$ /y on average. Thisyearly cost will decrease after 3-4 years in operations.

We plan to have this investment led by Switzerland andshared by an international consortium. This is essentialto ensure that the data centre design and activities

match the need and also benefit from the expertise ofthe broadest community. This also provides the solidityand stability required to support CTA over many years.For five years of operation, the Swiss participation willbe of the order of 5% of the total estimated constructioncost of CTA. Note that INTEGRAL operations may endclose to when CTA becomes operational, providing asmooth transition at ISDC.

Cost estimates

Page 15

0 M$

0.5 M$

1.0 M$

1.5 M$

2.0 M$

2.5 M$

3.0 M$

2008 2009 2010 2011 2012 2013 2014

Data Centre Manpower

Data Centre HardwaregAPD Camera Manpower gAPD Camera Hardware

Spending profile, assuming CTA operations will start in 2012

Manpower [FTE] Hardware [k$ /y]

Management and infrastructure

Policies, time allocation and science operations

Calibration, standard processing and science archive

Archival of other Cherenkov array data

Offline analysis and user support

On-site data centre activities

Raw data archive

Total

Development Operations Development Operations4.0 4.0 25 25

1.7 1.7 0 25

10.0 8.5 25 40

2.0 1.0 0 0

1.5 2.5 0 0

1.5 3.0 70 40

2.5 4.0 90 370

23.2 24.7 210 500

Cost estimates for the Cherenkov Science Data Centre


18/20


High energy astrophysics developed in Switzerland (mostly inGeneva and Zurich) since the 90’s at a sustainable rate byparticipating to international projects like XMM, RHESSI,INTEGRAL and MAGIC. During the same time a community ofscientists and engineers active in the domain has built-up from afew to about 40 people. This community is recognisedinternationally and is eager to participate in an expandingresearch field which matures in Europe through the CherenkovTelescope Array project.

By providing the science data centre for INTEGRAL, CTA and,

later, for XEUS, Switzerland brings a very valuable and concreteparticipation in these projects and plays a central role in manyaspects to foster the development of high energy astrophysics inEurope and in the world. Hosting the data centre for high energyastrophysics brings an important responsibility and visibility duringthe development, operations and post-operations of all thesescience missions. The worldwide scientific community relies on theservices provided and this permits scientists working inSwitzerland not only to participate in the science but also to takeresponsibilities that are central to the success of these experiments.

PSI has developed key technologies for high energy physics whichcould find an important application in the CTA cameras. Buildinga prototype camera will allow to qualify the technology andprepare industrialized mass production in view of building close toa million detectors for CTA.

Astrophysics at very high energies interest both the astronomicaland the particle physics communities. This cross fertilization willgrow further over the next decade and the location of both theCherenkov Science Data Centre and CERN in Geneva willcertainly help the process and provides many opportunities interms of joint activities, fostering of ideas and public visibility. The

very central place of Geneva in particle physics will be reinforcedand extended towards astro-particle sciences.

Perspectives

Impressum

The Cherenkov Telescope ArraySwiss Participation and PerspectivesNovember 2007

Contributions:

CTA FP7 design study proposalH.E.S.S. & MAGIC collaborations

ISDC, IPP, DPNC & PSI

Contacts:

ISDC# Dr Roland Walter # ([email protected])# Prof Thierry Courvoisier # ([email protected])

IPP# Dr Adrian Biland# ([email protected])# Prof Felicitas Pauss

# ([email protected])#

DPNC# Prof Martin Pohl# ([email protected])

PSI# Dr Dieter Renker # ([email protected])

Editior:

Roland Walter

INTEGRAL Science Data CentreChemin d’Ecogia 16, CH–1290 Versoix

Page 16


19/20


The central regions of the Milky Way, observed at soft gamma-rays by INTEGRAL : black holes and neutron stars

Sunset on the MAGIC segmented mirror; observations are about to start


20/20


iact swiss participation

Documents