Detector Considerations for a HAWC Southern Observatory
DDetector considerations for a HAWC southernobservatory
Michael DuVernois † for the HAWC Collaboration † Department of Physics and Wisconsin IceCube Particle Astrophysics Center (WIPAC),University of Wisconsin–Madison, Madison, WI, USAE-mail: [email protected]
The High-Altitude Water Cherenkov (HAWC) observatory in central Mexico is currently theworld’s only synoptic survey instrument for gamma rays above 1 TeV. Because there is signif-icant interest in covering the full TeV sky with a survey instrument, we have examined optionsfor a Southern Hemisphere extension to HAWC. In addition to providing all-sky coverage ofTeV sources, a southern site could complement existing surveys of the densest part of the Galac-tic Plane, provide continuous monitoring of Galactic and extragalactic transient sources in bothHemispheres, and simplify the analysis of spatially extended signals such as diffuse gamma raysand the TeV cosmic-ray anisotropy. To take advantage of the air-shower physics and lower theenergy threshold of the experiment as much as possible, a high altitude site above 5000 m a.s.l(vs. 4100 m a.s.l. at the current site in Mexico) has been specified. To facilitate efficient detectorconstruction at such altitudes, the detector tanks would be assembled at lower altitude and deliv-ered to the site. An all-digital communications and data acquisition scheme is proposed. Possibledesigns include taking advantage of digital optical module technology from the IceCube experi-ment as well as new custom electronics. We discuss the physics potential of such an experiment,focusing on the energy threshold, angular resolution, and background suppression capability ofthe experiment, as well as the advantages of full-sky coverage above 1 TeV.
Corresponding authors:
M. DuVernois ∗ The 34th International Cosmic Ray Conference,30 July- 6 August, 2015The Hague, The Netherlands ∗ Speaker. c (cid:13) Copyright owned by the author(s) under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike Licence. http://pos.sissa.it/ a r X i v : . [ a s t r o - ph . I M ] A ug AWC South
M. DuVernois
1. HAWC Observatory
The HAWC Collaboration has built a high-energy gamma-ray observatory at 19 degrees Northlatitude and 4100 m a.s.l. in the Sierra Negra–Pico de Orizaba saddle valley in the State of Puebla,Mexico. The array consists of 300 water Cherenkov detector (WCD) tanks instrumented withphotomultiplier tubes (PMTs) to detect and record incident gamma and cosmic ray air showers viathe Cherenkov emission produced as the shower particles traverse the tank water volume. The fulldetector was completed and inaugurated in early 2015 with the partial detector taking data duringconstruction.[2]The 300 tanks and 1200 PMTs in HAWC have an active collecting area of about 12,000 m over a total site area of about 22,000 m . HAWC can detect air showers with energies from about100 GeV to hundreds of TeV. The direction of the primary particle is reconstructed from the arrivaltime distribution in the array, the energy from the PMT charge signals, and composition (distin-guishing astrophysical photons from the much more abundant hadrons) from the shower topology.Charge deposited in the tanks is estimated by a pair of time-over-threshold (ToT) measurementsfor each PMT.The primary physics goals of the HAWC Observatory are the detection of new TeV gamma-raysources and the investigation of transient behaviors at TeV. The observatory complements all-skylower-energy coverage from the Fermi and SWIFT satellites and narrow-beam observations of thehigh-energy sky with Imaging Air Cherenkov Telescopes (IACT) such as MAGIC and VERITASin the Northern Hemisphere and HESS in the Southern Hemisphere. Each day HAWC surveysabout 2/3 of the sky from its low latitude location.The future of gamma-ray astronomy will be significantly different than the present with VER-ITAS scheduled for shutdown soon and the Cherenkov Telescope Array (CTA) collaboration look-ing to build IACTs in both the Northern and Southern Hemispheres. With HESS already operatingand CTA soon to be building in the Southern Hemisphere, a wide field of view instrument capableof acting as a survey instrument for the enhanced deep sensitivity of CTA for the southern skyseems more than sensible. To keep pace with the evolution from VERITAS to CTA, a similar in-crease in effective area and sensitivity from HAWC in the north (Mexico) to HAWC in the south(HAWC-South) is required.As an interim step for detector development, and also to enhance the ability of the HAWC arrayto pinpoint shower core locations when the cores are off the edge of the array, there is a plan toextend the current HAWC detector using a sparse array of “outrigger" tanks. The electronics for theoutriggers will be a development platform for HAWC-South electronics efforts. The outriggers aredescribed below and are a technological link, in both tank hardware and electronics, from HAWCto a potential larger, higher-altitude HAWC-South.
2. Outriggers
With the construction of the 300 HAWC tanks completed, and routine data-taking underway,the HAWC collaboration is exploring ways to increase the sensitivity of the detector. The compactarray of tanks in HAWC has a large perimeter/area ratio, and this large amount of ’edge’ meansthat a significant fraction of the high-energy showers that trigger the detector are located outside the2
AWC South
M. DuVernois main array. In the HAWC-predecessor experiment Milagro, this was mitigated with a thin samplingof small water tanks each with one PMT arrayed outside of the dense inner region. This allows foraccurate core and energy measurements even for events which fall somewhat outside of the innerregion of the detector. Simulations have shown that O ( ) small water tanks (perhaps commercialwater storage tanks which are ubiquitous in Mexico) with single PMTs in each could double theeffective area of the experiment.The HAWC detector uses time-over-threshold (ToT) electronics with PMT signals transmittedto a central counting house via analog cables. The electronics were partially inherited from theMilagro experiment and new modules can no longer be produced. In addition, it is not practical totransmit analog signals over the longer distances required in an outrigger array or larger detector.Therefore, distributed electronics are the preferred solution are the preferred solution, and a simpleelectronics layout is shown in Fig. 1. This scheme takes advantage of inexpensive, low-powerFADCs and FPGAs to digitize and then generate a total charge (integral) and time stamp (viasynchronous Ethernet) which mimics the data of HAWC, but through a very different path. AtHAWC, the data rate is between 25–50k hits/s over a dynamic range of 0.25 to 10k photoelectronswhich constrains the availability of commercial readout schemes. The full FADC is availablefor calibration data, or pre-scaled event readouts, while allowing for the simpler (and HAWC-compatible) time plus charge measurement.Fig. 2 shows the same block diagram as Fig. 1 in terms of selected, available parts. Thisset of electronics is undergoing prototyping, with lab versions expected operational in late 2015.The electronics are equivalent (total charge and arrival time) to the existing system at the triggerand analysis levels. However, the outrigger electronics system is more flexible, allowing for fullwaveforms on request, local histogramming, pre-scaling, and feature extraction. This developmenteffort has been taking advantage of overlapping parts and requirements with the IceCube Gen2DOM ([1]) development, for example, with common high voltage modules, shaping electronics,and similar programmable logic.Figure 1: Block diagram for the prototype outrigger electronics.This set of electronics, and work on small commercial water tanks, could inform the design ofthe HAWC-South detector which, for reasons detailed below, is likely to require smaller tanks thatcan be transported completely outfitted (save for the water) up the mountain, and distributed low-power (possibly solar-powered) electronics. Full waveform digitization is not ruled out however3 AWC South
M. DuVernois
Figure 2: Chip-level hardware selections for the principle components in the outrigger electronics.as the down-stream computing, which would be the major limiting factor right now, advances inrough accord with Moore’s Law.
3. HAWC-South
There are four general areas of improvement in capability for a Southern Hemisphere largearea water Cherenkov detector that are of interest in moving into the CTA-era of TeV gamma-raysensitivities. These gains would be in line with the sensitivity improvements of CTA over VERI-TAS, or put another way, a HAWC-South would need to be large enough to act as an interestingsurvey instrument for CTA.These areas of improvement are higher altitude, larger area, improved hadronic rejection, andimproved shower sensitivity. Although we’ll consider these one at a time, these issues are all tightlycoupled to each other: for example, higher altitude inherently gives more “ground level" particlesfor a given primary energy, effectively improving both the shower detection sensitivity and the totaleffective area.
Increasingly the altitude of the detector site is a huge win in terms of signal strength, lowerenergy threshold, and effective area, but subject to the constraints of both human physiology andland area available at a given altitude. At lower energies, the ability to cross-check with IACTsis enhanced. For a rough feel of the altitudes under consideration and the air shower profile, seeFig. 3. Sites at 5400 m are available in the Chilean Atacama Desert, for example the University ofTokyo Atacama Observatory is being built at 5640 m and a number of nearby 20,000 m hilltopsnearby are at altitudes between 5200 and 5400 m. The well-known cosmic-ray observatory atChacaltaya in Bolivia is above 5200 m as well. (See [3] for a sensible starting point to look forexisting astronomical observatories as potential Southern sites.)At these altitudes, freezing of the detector water and the inability of unadapted personnel towork are major problems. Insulation and passive solar gain on the tanks are a potential solution tothe former issue. For site-work, it seems likely that the detector (water tank, PMT, and electron-ics) would need to be constructed at lower altitude and then delivered to the high altitude site forlater filling. Roads to the high-altitude Atacama sites appear to be reasonable; see Google Earth’simagery of the University of Tokyo Atacama Observatory or the ALMA Observatory sites.4 AWC South
M. DuVernois a t m o s p h e r i c s l a n t d e p t h [ g c m − ] MilagroHAWC5400 m h e i g h t a b o v e s e a l e v e l [ k m ] Figure 3: Air shower development with Milagro, HAWC, and potential 5400m site.Simulations of a strawman detector were performed for a variety of altitudes (4100 m forHAWC in Mexico, 5200m for plausible sites, 6000 m for maximum Southern altitudes of flat sites,and 6600 m for an extreme reference). See Fig. 4 for the ground level energy of gamma-rays below1 TeV. The simulated detector was a 30 ×
30 array of 1.5 m radius and 1.5 m tall tanks separated by3 m and with a single 10" high-quantum efficiency PMT in the water. This detector yields estimatedCrab Sensitivities as shown in Fig. 5 with the experiment sited at 20 ◦ South.Any detector at higher altitudes will have to deal with higher rates. This could be amelioratedwith a larger number of smaller tanks, local coincidence requirements, or significantly faster triggerprocessing.
A straightforward path to improved photon sensitivity is simply increasing the area of thedetector array. This would need to be carefully matched to the available high altitude sites, whichat above 5200m seem to be constrained to less than 20,000 m of roughly level ground beforedropping down the slopes of the hill. Likely the best route to increased area would be via lower5 AWC South
M. DuVernois
Figure 4: Surviving energy below 1 TeV.per-tank costs of construction and instrumenting a couple of closely spaced hilltops with a commonacquisition system.
Improvements in the gamma-hadron separation directly yield improved gamma-ray rates anddecreased backgrounds. With a single detector technology (the water tanks), hadron rejectioncomes from the shower topology. That is, hadronic cascades generate sub-shower clumps of de-posited energy whereas photon showers show only the smoother electromagnetic component. Di-rect detection of muons and nuclear fragments in the air showers could also be used for gamma/hadronseparation. Muon detection in the water Cherenkov tank, underground muon paddles, or dividingthe tank into top (electromagnetic) and bottom (hadronic) segments are all options in this direction.All would effectively double the cost per tank, and the trade-off between increased numbers oftanks and increased background suppression has not yet been fully studied.Other ideas in this direction include thin layers of liquid scintillator in thin bags within thetank, read out by a small PMT, or boron-laced plastic scintillators which observe the neutron back-splash from the Earth shortly after a hadronic shower touches down. These techniques are not yetproven in large detectors, but can also be prototyped at the HAWC Observatory in Mexico.
Improving the sensitivity of an individual single water tank implies either increasing the lightyield in the water and/or collecting more of the photons produced in the tank. The Cherenkovlight yield is difficult to modify but adding in a scintillation component is possible with a mix ofwater and liquid scintillator (which could help prevent freezing as well). More photon collectionis possible with a web of wavelength-shifting fibers dispersed through the water and light-piped toPMTs. With the high altitude of the potential sites, the amount of “field-work" at the site will need6
AWC South
M. DuVernois
Figure 5: Sensitivity of a HAWC-like array to a gamma-ray source with the same spectrum as theCrab Nebula as a function of source declination and detector altitude.to be kept to a minimum, so designs like this would need to be built down the mountain and takenup largely intact.
4. Additional technologies
In efforts to increase either the total photon yield at the PMTs, or enhancing the discriminationbetween the background hadrons and the signal gamma-rays, ideas such as placing scintillatorsabove or below the tanks, adding dopants to the water, segmenting the tanks top and bottom, oradding neutron detectors (boron-doped scintillators) under the tanks are under consideration. Envi-ronmental concerns and the location of the HAWC site in Mexico, which is in a national park, havemade buried detectors or dopants rather difficult to test with the existing detectors. At the HAWCsite, the installation of prototype electronics and different technology detectors is relatively easyand their operation in concert with the main detector can be used for cross-calibration.To more adequately address the testing of prototype detectors and novel detector technologies,two tanks near the counting house will be outfitted with more general readout electronics, and havetime-tagging which is compatible with the HAWC time stamps but also easily available to newdetectors. IceCube electronics, specially modified to handle the high data rates at the site, will bethe first electronics tested in these tanks. A flexible interface standard is being designed to allowfor easier access to interface with the HAWC data flow.7
AWC South
M. DuVernois
5. Schedule and prospects
Outrigger electronics development is underway at the University of Wisconsin during 2015with a goal of a deployable sample readout system available Spring 2016 for one of the water tanksat HAWC in Mexico. Large-scale deployments might be possible later in 2016. Parallel worktaking advantage of the IceCube distributed digitization electronics is also underway with a similarprototyping time scale. Visits to potential HAWC South sites are also ongoing. With the immensepotential synergies between a Southern Hemisphere large area water Cherenkov detector and theplanned Southern CTA telescopes, further cooperation between the groups is foreseen.
6. Conclusions
The HAWC Observatory is in routine operations now, with data flowing to collaborators in theUSA and Mexico and early publications are appearing (see [2] for an overview of current results).Design work is ongoing for a new set of electronics and smaller water tank setups for the outriggersintended to go around the inner core of detectors at HAWC. This design work should also help tobridge the gap between the time-over-threshold electronics located in a central counting house inHAWC (and partially inherited from the Milagro experiment) and distributed waveform digitizationwith local pre-analysis in a future HAWC-South design. Custom electronics are required for thisdue to the very high singles rate at HAWC altitudes.There are plausible southern hemisphere sites for a detector that would be well-matched asa “finder-scope” paired to the CTA sensitivity and directionality. The sites are in South Americawhich could work well with CTA potential sites. Alternate detector technologies to replace or aug-ment the water Cherenkov technique are also being investigated, but would need to be sufficientlylow risk to become baseline plans.
References [1] M. A. DuVernois et al. (IceCube Gen2 Collaboration, The IceCube Gen2 Digital Optical Module andData Acquisition System, Proc. of the 34th ICRC, The Hague, 786 (2015).[2] J. Pretz et al. (HAWC Collaboration), Highlights from the High Altitude Water CherenkovObservatory, Proc. of the 34th ICRC, The Hague, 866 (2015).[3] https://en.wikipedia.org/wiki/List_of_highest_astronomical_observatorieshttps://en.wikipedia.org/wiki/List_of_highest_astronomical_observatories