Upgrade of the VERITAS Cherenkov Telescope Array
aa r X i v : . [ a s t r o - ph . I M ] J u l PROCEEDINGS OF THE 31 st ICRC, Ł ´OD ´Z 2009 1
Upgrade of the VERITAS Cherenkov Telescope Array
A. Nepomuk Otte ∗ for the VERITAS collaboration †∗ Santa Cruz Institute for Particle Physics, University of California at Santa Cruz, CA 95060 Santa Cruz, U.S.A.([email protected]) † see R. A. Ong et al. (these proceedings) or http://veritas.sao.arizona.edu/conferences/authors?icrc2009 for a full author list Abstract . The VERITAS Cherenkov telescope arrayhas been fully operational since Fall 2007 and hasfulfilled or outperformed its design specifications.We are preparing an upgrade program with thegoal to lower the energy threshold and improve thesensitivity of VERITAS at all accessible energies. Inthe baseline program of the upgrade we will relocateone of the four telescopes, replace the photo-sensorsby higher efficiency photomultipliers and install anew trigger system. In the enhanced program of theupgrade we foresee, in addition, the construction ofa fifth telescope and installation of an active mirroralignment system.
Keywords : Cherenkov telescope, IACT, VERITAS,photon detectors, trigger, gamma ray astronomy
I. I
NTRODUCTION
Gamma-ray astronomy with air-shower imagingCherenkov telescopes (IACTs) is a young discipline.With the latest generation of instruments much insighthas been gained into the non-thermal processes in a vari-ety of astrophysical sources, such as, for example, activegalactic nuclei, supernova remnants, and pulsar windnebulae. Observations in gamma-rays above 100 GeValso provide a tool to study interesting questions that areof fundamental importance in physics and cosmology.Some examples are the nature of dark matter and thedensity and evolution of the extragalactic backgroundlight.Of the existing IACTs the VERITAS Cherenkovtelescope array is one of the most sensitive gamma-ray instrument above 100 GeV. VERITAS can detecta point source with a power law spectrum of index -2.5 and 1% of the Crab Nebula flux in less than 50hours at a significance level of σ . The angular andenergy resolutions are energy dependent. At 1 TeV thereconstructed arrival direction for 68% of all gamma-ray events fall within ∼ . ◦ of the true arrival direction(cf. Figure 1); the energy resolution at this energy is ∼ .The VERITAS array has achieved, and in somemetrics surpassed, its design specifications and sciencegoals; however, the performance parameters are not lim-ited by any fundamental physical constraints. Improvedperformance can be achieved through improved data Fig. 1. Angular resolution as function of energy. ”standard” cutsrefer to gamma-ray selection cuts suitable for sources with an energyspectrum similar to or harder than the Crab Nebula. “soft” cuts areused for sources with correspondingly softer spectra. analysis techniques and hardware upgrades, which willallow us to better address the key science goals.In this paper we describe a staged upgrade programfor VERITAS, to envisaged over the next few years, withthe primary goals to lower the energy threshold of theinstrument and to increase its sensitivity; both of thesewill allow us to better address our key science questions.We describe a baseline program, which includes: • relocating one of the telescopes, • refitting the telescope cameras with more sensitivephoto-detectors, and • implementing a different hardware trigger scheme.A more ambitious enhanced program is also describedwhich, in addition to the baseline upgrades, involvesthe construction of an additional telescope and thedevelopment of an active mirror alignment system.II. B ASELINE P ROGRAM
The baseline program includes the relocation of onetelescope, an upgrade of the focal plane instrumentationwith high efficiency photon detectors, and a new trigger
OTTE et al.
VERITAS UPGRADE system. The first step, the relocation of telescope T1 ispresently ongoing.
A. Relocation of Telescope T1
Fig. 2. A plan view of the Whipple Observatory Basecamp, takenduring the construction of telescope 4. The four existing VERITAStelescope positions are indicated, along with the planned position oftelescope 1 after relocation and a potential site for a new telescope,T5.
The VERITAS array is located at the basecamp ofthe Whipple Observatory in southern Arizona. The firsttelescope was placed there in 2002-2003 with the an-ticipation to relocate the array to a different location. Itwas subsequently decided to build the whole array atthe same site, which resulted in the current layout thatis shown in Figure 2. With a baseline of 35 m, T1 andT4 allowed some interesting studies of the effect thatthe close proximity of two Cherenkov telescopes hason the sensitivity of the array around 100 GeV [1]. Atthe same time, the proximity of T1 and T4 results in asuboptimal effective collection area at higher energies,which can be maximized by a more uniformly shapedrectangular configuration of the array.Detailed simulations have shown that the sensitivity ofthe array increases by 15-20% (cf. Figure 3) if telescopeT1 is relocated to the position indicated in Figure 2. Themost important practical result of this relocation is thatthe time required to detect a weak source is reduced byabout 30%, which is equivalent to extending the possibleobservation time from an annual 1000 hours to 1400hours annually.The relocation of T1 is presently under way andexpected to be finished by the beginning of the upcomingobservation cycle in fall 2009.
B. Higher Efficiency Photon Detectors
We have considered several possible upgrades for theVERITAS telescope cameras: • an extension of the field of view, • a smaller pixellation, and • improved photon collection efficiency. ] -1 Gamma-ray Rate [min4 5 6 7 8 9 R e l a t i ve S e n s i t i v i t y VERITAS-2008VERITAS-A
Fig. 3. Effect of the relocation of Telescope 1 on the sensitivityas a function of the gamma-ray rate after cuts, normalized to thepeak sensitivity for the current layout of the array. “VERITAS-2008”and “VERITAS-A” refer to the configuration before and after therelocation, respectively.
These are all desirable improvements but the most ef-fective, practical, and economical upgrade is to increasethe photon collection efficiency by replacing the existingPMTs with higher quantum efficiency devices that havebecome commercially available. These PMTs have peakphoton detection efficiencies of about 35% in the blue,as compared to a peak efficiency of 18% of the devicescurrently in use on VERITAS.
Fig. 4. Effective area of VERITAS after standard analysis; dots forthe current configuration and squares for the expected improvementafter an upgrade with high efficiency PMTs.
Replacing the photon detectors with these higher effi-ciency “super bialkali” photomultipliers would result inan increase of the optical throughput for Cherenkov lightfrom an air shower by about 40%. Our simulations showthat the immediate effect of this upgrade would result inabout 17% improvement in the sensitivity of the arrayfor a weak point source above 100 GeV. Of possiblyhigher importance is the increase of sensitivity of thearray at and below 100 GeV where the effective area isexpected to increase by about one order of magnitudeif our standard analysis tools are applied (see Figure 4).If an analysis tuned for higher sensitivity at 100 GeVis used we expect an overall significant increase of the
ROCEEDINGS OF THE 31 st ICRC, Ł ´OD ´Z 2009 3 effective area at 100 GeV.
C. A Topological Trigger
Additional improvement of the performance at lowerenergies is expected by implementing a more intelli-gent hardware trigger system as part of the baselineVERITAS upgrade program. The trigger threshold inCherenkov telescopes is largely dependent on the abilityto suppress accidental triggers caused by fluctuationsin the night-sky background light (NSB) and triggerscaused by cosmic rays, mostly protons, muons, andelectrons. While stereoscopic operation already providesexcellent suppression of background because of tighttrigger coincidence windows of 100 ns or less betweentelescopes, we expect further improvements as the resultof two parallel trigger system research and developmentefforts that have been going on over the past few years.In one effort we developed an FPGA based simplethree-fold neighbor logic that achieved about 4 ns re-solving times in a test on one of the telescopes wherethe system was running for one month in 2008. A 4 nscoincidence window would allow us to reduce the triggerrate caused by fluctuations in the NSB by more than afactor of two.In another effort, we are following a different ap-proach to develop a three-stage, high-speed trigger tosuppress accidentals from both the NSB and cosmicrays. This “topological trigger” [4], [2] consists of high-speed camera-level triggers that transmit image parame-ters to the array trigger for real-time event classification.The topological trigger is a shift from off-line to real-time gamma/hadron separation.III. E
NHANCED P ROGRAM
A. Array Expansion
An enhanced upgrade option would be the construc-tion of a fifth telescope, with the obvious impact of anincrease in the effective area. This increase would in turnimprove the sensitivity by an amount approximately pro-portional to the square root of the number of telescopesin the array.In reality, an improvement in sensitivity larger than12% is expected because of the better angular resolu-tion [3] and better gamma-hadron separation. A betterangular resolution will allow for better suppression ofbackground events and will have a bigger impact onmorphological studies of Galactic sources and the studyof possible gamma-ray of halos around extragalacticsources. Another advantage of a five-telescope arraywould be the operation in split mode, e.g. the parallel ob-servation of two sources with two and three telescopes,respectively.Detailed studies with Monte Carlo simulations areunder way to investigate the various aspects and impactsthat the addition of a fifth telescope would have.
B. Active Mirror Alignment
Each VERITAS telescope has 350 mirror facets. Arealignment becomes necessary every time a facet isrecoated, which is done approximately every two years.The alignment procedure is done manually [5], whichis manpower intensive and better done in an automatedway. An automated alignment procedure would alsocorrect for elevation-dependent mechanical deformationsof the optical support structure. This task can be accom-plished by an active mirror alignment system (AMAS).An AMAS would eliminate the aforementioned disad-vantages of the alignment method currently used and, inaddition, would allow monitoring the reflectivity of indi-vidual mirror facets [6], changing the focus from infinityto the height of the shower maximum, and altering theoptical axis of the OSS to compensate systematic effects,for example due to camera sagging. Finally, the abilityto defocus the OSS provides an additional safety featurein the case of a telescope drive failure.IV. D
ISCUSSION
The VERITAS gamma-ray observatory in its presentconfiguration has achieved its design specifications andis one of the most sensitive gamma-ray instruments. Inthis paper we have outlined an upgrade program, whichwould allow us to increase the sensitivity and lower theenergy threshold of VERITAS.In summary the upgrade program comprises: • relocating one of the VERITAS telescopes, • upgrading the VERITAS cameras with photon de-tectors of higher efficiencies, and • implementing a novel trigger system.The enhanced upgrade program includes all of the aboveoptions plus • adding a telescope, and • automatic mirror alignment system.The discussed upgrade options will result in a sig-nificant improvements of the performance of VERITASand ultimately on all science topics that we addresswith VERITAS. A lower energy threshold will allowus to look deeper into the Universe, and at the sametime increases the overlap in energy with the gamma-ray satellite Fermi.A higher sensitivity at all energies will provide uswith the possibility to observe more sources or performdeeper observations of particular objects. The improvedsensitivity after the relocation of telescope T1 is in partdue to a better angular resolution, which will also allowus to make morphological studies with higher resolutionsthan presently possible.Of all the discussed upgrade options the move of T1has been fully funded and is presently ongoing. All otherupgrade options are still under discussion with parallelefforts in simulations and research and development. OTTE et al.
VERITAS UPGRADE
V. A
CKNOWLEDGMENT
This research is supported by grants from the USDepartment of Energy, the US National Science Foun-dation, and the Smithsonian Institution, by NSERC inCanada, by Science Foundation Ireland, and by STFCin the UK. We acknowledge the excellent work of thetechnical support staff at the FLWO and the collaborat-ing institutions in the construction and operation of theinstrument. R
EFERENCES[1] J. Kildea et al., “Development of a Low-energy Trigger forVERITAS”, 2009, in “High Energy Gamma-Ray Astronomy:Proceedings of the 4th International Meeting on High EnergyGamma-Ray Astronomy”, Eds F. A. Aharonian, W. Hofmann &F. Rieger, AIP Conf. Proc., , 760[2] J. Anderson et al., “A New High-Speed Pattern RecognitionTrigger for Ground-Based Telescope Arrays Used in Gamma RayAstronomy”, 2008, Nuclear Science Symposium Conference,IEEE, 2773.[3] S. Funk & J. A. Hinton, “Simulation studies of the high-energycomponent of a future imaging Cherenkov telescope array”,2009, in “High Energy Gamma-Ray Astronomy: Proceedingsof the 4th International Meeting on High Energy Gamma-RayAstronomy”, Eds F. A. Aharonian, W. Hofmann & F. Rieger,AIP Conf. Proc.,1085