HiSCORE and TAIGA

The origin of cosmic rays

Since the first detection of cosmic rays early in the 20th century there have been many efforts to try to find out how and where they are produced. An intrinsic problem in that process is that cosmic rays consist of charged particles and are therefore deflected in interstellar and intergalactic magnetic fields, which makes it impossible to track down their origin.

It is possible, though, to measure the energy spectrum and the composition of cosmic rays, and it has been done so in many experiments. Above a few GeV, the spectrum can be described by a powerlaw shape with changes of the spectral index at 4 PeV (the knee), at 400 PeV (the second knee), and at 1 EeV (the ankle). More than 99 % of all cosmic rays are hadrons, mainly protons and helium nuclei.

To examine the question of the origin of cosmic rays, two principal ways have been suggested. Firstly, neutrinos are expected to be produced and accelerated by decay of high energy mesons in the vicinity of cosmic ray accelerators. As neutrinos are not charged and do in fact interact very little at all, they are perfectly suited as messengers. However, detection of cosmic neutrinos is challenging due to their small interaction cross section, and all efforts in that direction haven't led to much sucess so far. Also, if anything will be detected, the signals would be weak and a detailed study (of spectra, for example) may not be possible.

The second possible way to explore the origin of cosmic ray is the search for high energy photons (i.e. gamma rays) that are produced in the same processes. Both direct (satellite based) and indirect (ground-based) methods have yielded in many interesting results during the last years. While satellite based experiments are excellent in detecting relatively low energy gamma rays, their area is too small to collect enough photons at higher energies (as those are much rarer). Photons with energies above 100 GeV can be detected with current Cherenkov telescopes (see link section).

Current telescopes such as H.E.S.S have effective sensitive areas of about 105 m2 which enables them to collect enough statistics up to about 10 TeV. The planned Cherenkov Telescope Array (CTA) will both lower the threshold below 100 GeV and, by its larger instrumented area, raise the upper energy limit far above the capability of current systems. Nevertheless, it will probably not be affordable to instrument areas far above one square kilometer with Cherenkov Telescopes.

HiSCORE is being designed in order to access energies beyond the range of CTA. The use of simple, inexpensive detector stations, which are placed several hundred meters apart, makes it possible to instrument a huge area. The energy threshold will be around a few 10 TeV and it will be perfectly suited for energies from 100 TeV up to at least 1 EeV (see simulation section). With it, it will be possible to extend the energy range of gamma ray spectrums of many known sources to higher energies which will help to adress many interesting questions.

Currently detected gamma ray sources can seldom be identified for sure as accelerators of cosmic rays, as gamma rays may be produced by other processes as well. In the leptonic scenario it is assumed that the gammas are accelerated by high energy electrons via the inverse Compton effect. This effect, however, becomes inefficient at certain, higher, energies (Klein-Nishina regime). Examing the higher part of energy spectra of gamma ray sources should therefore settle the ambiguity between the leptonic and the hadronic scenario. If a source can is found to emit gamma rays at energies in the multi-TeV range, it would be a strong indication of it being an accelerator of high energy cosmic rays.

Galactic photon background

If spectra of sources are assumed to be known, the distortion of the spectrum by cosmic absorption can be used to infer information about the galactic photon fields.

Large structures

As HiSCORE offers a wide field of view it is perfectly suited to monitor extended gamma ray emitting structures such as molecular gas clouds, dense regions or large scale structures such as star forming regions or the galactic plane. Gamma-ray signals are expected from the interaction of the molecular gas with charged cosmic rays trapped in our Galaxy (meson decay). Spectral analysis of the faint diffuse gamma-ray signal correlated with the gas density can be used to derive the cosmic ray energy density at the clouds and thus map cosmic ray accelerator properties and propagation throughout the galaxy.

Here again, observations at the highest energies with a wide field of view are required. Recently, MILAGRO was able to detect largely extended gamma-ray emission from the Cygnus region. The successor experiment HAWC will improve the sensitivity and lead to further progress here. However both experiments are optimized for a low energy threshold and it will be difficult to achieve the necessary sensitivities in the cut-off regime of cosmic accelerators.

Proton-proton cross section at high energies

As HiSCORE offers the possibility to reconstruct the altitude of the shower maximum and therefore the height of the first interaction of a primary proton, the interaction cross section between protons can be measured. The energy range of HiSCORE offers to investigate center-of-mass energies between 1 TeV and 100 TeV, therefore extending greatly the range of the measurements at the LHC.

Dark Matter Search

The spectrums of gamma ray sources are distorted by the absorption of gamma rays in the interstellar photon fields and the CMB. A strong absorption feature is expected around 100 TeV. If the spectrum of the source is assumend to be known, examination of the absorption can bring information about the interstellar photon fields. On the other hand, photons might be converted into axions or hidden photons along the way, particles which might travel long distances without being affected by normal absorption processes and the convert back into normal photons. If the absorption is less than expected, this might be an indication for the presence of hidden photons or axions. Also, the decay of heavy supersymmetric particles might lead to a detectable signal in HiSCORE. Therefore, HiSCORE can also be used for testing theories about Dark Matter.

As many other experiments in the field (see Links section) the HiSCORE detector will make use of the fact that high energy particles (both photons and hadrons) that hit earth's atmosphere produce a so called air shower consisting of many secondary particles.

Due to the high energy of the primary particle, many of these secondaries have kinetic energies that make them faster than the speed of light in air, therefore emitting cherenkov light. This light is weak, but can be detected on the surface of the earth with sensitive instruments.

The HiSCORE Detector is planned to consist of a number of detector stations with a spacing of more than hundred meters and is designed to sample the Cherenkov light front. Each station will consist of four modules or channels.

Each module consists of one large area photomultiplier looking upwards and the respective electronics and readout components. To increase the effective sensitive area of the detector, a Winston Cone will be placed on top of the PMT, concentrating the incoming light (see Fig 2 and 3). A useful side effect of the Winston Cone is the reduction of stray light from the horizon and the Night Sky Background (because photons with angles larger than a certain cutoff angle, say 30 deg, are not transmitted onto the PMT).

For a cutoff angle of 30 degrees the ratio of R1 and R2 must be about 2. At the moment, it is planned to use 20 cm (8 inch) PMTs as offered e.g. by Electron Tubes. In this case one module would have a sensitive area of about 0.125 sqm.

If one would use, for example, four modules per station, each station would have an effective sensitive area of about half a square meter. Currently simulation studies are ongoing to determine the optimal size of the detector stations.

Currently the construction and testing of a prototype module is underway. Different photomultiplier modules are being tested in a test-bed and a slow-control concept was developed and implemented. We plan to soon start running a prototype station in Hamburg using scintillator material (for test purposes) inside the station. A 3D-schematics of the planned station mechanics is shown in Figure 4.

The first prototype Winston cone has been assembled by the workshop at the University of Hamburg. The Winston cone material is a reflective foil manufactured on a plastic-like support, easily cut and bent, and therefore ideal for implementation in the barrel-mount concept (Fig 6). Several Electron Tubes 8'' photomultipliers and Hamamatsu models with 8'' and 10'' are available and are being tested in our test-bed.

An event reconstruction has been developed (see Documents section for detailed description) which reconstructs key parameters of the air shower as shower core position, shower depth, energy and direction of the primary particle. By comparing the reconstructed values to the Monte Carlo input parameters, the resolution of the detector can be estimated.

It has been found that a basic gamma/hadron separation can be applied by using timing information. Currently, the power of self-learning algorithms for this task is under study.

An important value to describe the power of a gamma ray instrument is the point source sensitivity. It is defined as the minimal gamma flux that is needed to identify a source (with more that 5 sigma above the Cosmic Ray background) in a certain observation time. For Cherenkov Telescopes usually a time of 50 hours is used, while sensitivities of wide angle instruments like HiSCORE or HAWK are usually given for a longer time period, as they can watch all sources in the field of view at the same time.

Shown in Fig 1 is a preliminary sensitivity plot for SCORE (10 sqkm) and HiSCORE (100 sqkm) for 500 hours of observation and a miminmum of 50 events (which defines the high energy end of the sensitive energy range). For comparison, other currently existing and planned instruments are shown, too.

A new sensitivity curve, taking into account new results from the event reconstruction tests, will be posted here soon.

In Fig 2 a skymap in galactical coordinates is shown with a colour map, indicating possible observation times for the various regions of the sky. Given are the observation times for one year (about 1500 hours of observation). Additionally, all so far known TeV gamma ray sources (black dots) and the supergalactic plane (black line) are plotted.

The simulation takes into account the position of the moon (and the sun, of course), as only moonless nights are suitable for observation. In reality, a moderate decrease of about 25% of the shown observation times is expected as the duty cycle is reduced by bad viewing conditions (clouds, mainly).

In this version of the the skymap a detector position at about 31 deg south is assumed (southern Australia, Namibia, ...).

As one can see, a good number of known TeV gamma ray sources are within the field of view of HiSCORE, as well as a large part of the galactic plane (horizontal line at zero degrees) and a part of the supergalactic plane.