2.2. Ground-based Telescopes

The Earth's atmosphere is as opaque to photons of energy above 300 GeV as it is to photons in the CGRO range (100 keV-30 GeV). However, at these higher energies the effects of atmospheric absorption are detectable at ground level, either as a shower of secondary particles from the resulting electromagnetic cascade or as a flash of Cerenkov light from the passage of these particles through the Earth's atmosphere. As in the CGRO range, -ray observations are severely limited by the charged cosmic particle flux which gives superficially similar signals at ground level and, for a given photon energy, is 10,000 times more numerous. It is not possible to veto out the charged cosmic-ray background with an anticoincidence shield. As such, it might seem impossible to do -ray astronomy with such indirect techniques. However, there are small, but significant, differences in the cascades resulting from the impact of a photon and a proton on the upper atmosphere; also, the electromagnetic cascade retains the original direction of the photon to a high degree, and the spread of secondary particles and Cerenkov photons is so large so that a simple detector can have an incredible (by space-based -ray detector standards) collection area.

In practice, the most successful detectors are atmospheric Cerenkov imaging telescopes (ACITs), which record the images of the Cerenkov light flashes and which can identify the images of electromagnetic cascades from putative sources with 99.7% efficiency (Aharonian & Akerlof 1997; Ong 1998). Originally proposed in 1977 (Weekes & Turver 1977), the technique was not demonstrated until 10 years later when the Whipple Observatory 10 m reflector (Fig. 1) was equipped with a primitive imaging camera and used to detect the Crab Nebula (Weekes et al. 1989). The technology was not new (arrays of fast photomultiplier tubes in the focal plane of large optical reflectors with readout through standard fast amplifiers, discriminators, and analog-to-digital converters), but the technique was only fully exploited in the past decade. Compared with high-energy space telescopes such as EGRET, ACITs have large collection areas (> 50,000 m²) and high angular resolution (~ 0°.1). ACITs also have reasonably good energy resolution (~ 20%-40%), but small fields of view (FOV) (< 5°) and a background of diffuse cosmic electrons which produce electromagnetic cascades identical to those of -rays. ACITs also have low duty cycles (< 10%) because the Cerenkov signals are faint and produced at altitudes of several kilometers, requiring cloudless, moonless skies for observations. Most of the results reported to date have been in the energy range 300 GeV-30 TeV.

Figure 1. The Whipple Observatory 10 m imaging atmospheric Cerenkov telescope

In recent years VHE -ray astronomy has seen two major advances: first, the development of high-resolution ACITs has permitted the efficient rejection of the hadronic background, and second, the construction of arrays of ACITs has improved the measurement of the energy spectra from -ray sources. The first is exemplified by the Whipple Observatory 10 m telescope with more modern versions, the Cherenkov Array at Themis (CAT; a French telescope in the Pyrenées; Barrau et al. 1998), and CANGAROO, a Japanese-Australian telescope in Woomera, Australia (Hara et al. 1993). The most significant examples of the second are the High Energy Gamma Ray Astronomy (HEGRA) experiment, a five-telescope array of small imaging telescopes on La Palma in the Canary Islands run by an Armenian-German-Spanish collaboration (Daum et al. 1997), and the Seven Telescope Array in Utah, which is operated by a group of Japanese institutions (Aiso et al. 1997). These techniques are relatively mature, and the results from contemporaneous observations of the same source with different telescopes are consistent (Protheroe et al. 1997). Vigorous observing programs are now in place at all of these facilities. A vital observing threshold has been achieved whereby both Galactic and extragalactic sources have been reliably detected. Many exciting results are anticipated as more of the sky is observed with this present generation of telescopes.

The atmospheric Cerenkov imaging technique has now been adopted at a number of observatories whose properties are summarized in Table 1.

Above 30 TeV there are enough residual particles in the electromagnetic cascades that they can be detected at high mountain altitudes using arrays of particle detectors and fast wave front timing (Ong 1998). These arrays have large collection areas (> 10,000 m²), good angular resolution (~ 0.5°), moderate energy resolution (~ 100%), good duty cycle (100%), and large FOV (~ 2 sr); however, their ability to discriminate -rays from charged cosmic rays is severely limited. Despite the early promise of these experiments, which led to a considerable investment in their construction and operation, no verifiable detections have been reported by particle air shower arrays.