IACTs are detectors for high-energy gamma quanta, installed on the surface of the earth. The name imaging atmospheric Cherenkov telescope (IACT) may not be inspiring, but contains most of the characteristics of this type of instrument:

• The detectors used have a light collection mirror and a camera, so they resemble optical telescopes at least superficially (but read on to see also the differences);
• These telescopes detect light produced by the Cherenkov effect, a radiation emitted by relativistic particles when being slowed down; the slowing down is in the atmosphere, where the high-energy gamma quanta get absorbed.
• IACT-s record many Cherenkov photons for a single original gamma; they are seen by the camera as an image whose characteristics allow to identify the recorded particle as a gamma, and to specify its direction and energy.

The MAGIC telescope project is a continuation of work that has been started before 1987 (with the HEGRA detector), both experiments targeting the observation of highly energetic photons from sources within and outside our galaxy, by terrestrial telescopes. These photons are not directly observable: because tha atmosphere has only narrow windows for wavelengths to pass, and high-energy gamma rays do not reach the earth. They get absorbed in the atmosphere, transparent essentially only for visible and infrared light, and for long radio waves, as shown in the diagram below for the entire electromagnetic spectrum.

The absorption of the primary gamma leaves behind an avalanche, called an electromagnetic shower. The numerous secondary charged particles in such a shower, for an incident gamma rather exclusively electrons and positrons, all radiate low-energy (visible to ultraviolet) photons, the Cherenkov radiation. The radiation is emitted at a characteristic angle with the radiating particle, an angle which widens as the atmosphere thickens. Most of the shower development happens at an altitude above sea level from 20 to below 10 km. The radiated photons have an energy corresponding to a window of penetration, and arrive in large enough numbers on the surface of the earth to become an indirect image of the shower, allowing identification against backgrounds and reconstruction of the original particle's direction and energy. Those familiar with high-energy physics instruments can consider the atmosphere as an unbounded and changing total absorption calorimeter, and the Cherenkov radiation observed as part of shower leakage - difficult conditions indeed.

The showering process and the generation of Cherenkov light in a foreward cone have two immediate experimental consequences: the light is spread over a large area, typically a circle with a diameter of 250m, and hence the light intensity per unit area on ground is low. This allows detection of a gamma impinging anywhere inside this disk, i.e.an effective area of 30 to 100 000 sq.m., as long as the initial energy is high enough to produce enough Cherenkov light. Conversely, the signals are weak, marginally detectable; hence, the instrumental sensitivity must be pushed as far as possible: the collection area (mirror surface) must be maximized, and the camera elements (photomultipliers) must respond to single photons with high efficiency. To further improve sensitivity, experiments are installed on mountain tops far from background light and with as little observation time lost due to clouds as possible.

The sketches on the left show an electromagnetic shower developing in the atmosphere, and the image it causes in a telescope whose axis is aligned with the shower axis, and which is inside the shower cone (a circle of some 250 m diameter on the surface of the earth).

Gammas of the high energies that can be recorded by IACTs are relatively rare events. They have to be discriminated against a cosmic ray background several orders of magnitude more abundant. These are mostly protons or light ionized atoms, producing (more dissipated) hadronic showers, in which the charged particles also radiate Cherenkov photons. Hadronic showers do not typically come from the direction in which the telescope is trying to observe a gamma source. Also, hadronic showers are much less concentrated; the hadrons interact via the strong interaction, producing hadrons and leptons as secondary particles; multiple electromagnetic and hadronic secondary showers appear, with large fluctuations in relative energy, spread over a volume much larger than for an electromagnetic shower. Hadronic showers frequently contain long-lived high-energy muons, whose radiation shows typical patterns and in some cases may also be mistaken for gammas. The image hadrons produce in the detector, therefore, has characteristics different from gamma shower images; using suitable discrimination algorithms, fairly clean gamma signals can be obtained. For a selection of typical events, see some of them as MAGIC has recorded them, in sequence 1 (tagged) or in sequence 2 (untagged) .

Compared to optical telescopes, IACTs have to make do with very sparse data indeed. A typical high-energy gamma shower will be a very short and weak flash of light, lasting few nanoseconds and recording of the order of one hundred Cherenkov photons as an image. After applying selection criteria, that image corresponds to a single observed gamma quantum.

Left, you can see an optical image of the Crab nebula, remnants of a supernova which was observed in China in 1054, in the constellation Taurus. Visible is a structured cloud of hydrogen gas, about 6000 lightyears from our solar system and today some 10 lightyears across. The picture was taken by the 3.5m WIYN telescope on Kitt Peak, Arizona. A typical exposure time is of the order of minutes.

In the gas cloud a source has been observed (in satellite experiments) to emit pulsed X-ray radiation (keV to MeV range), at a frequency of 30 Hz; this is believed to originate from a pulsar, a spinning neutron star, in the center of the nebula. The Crab also emits VHE gamma radiation, at a fairly constant rate; this radiation has been observed by all existing IACTs, also HEGRA (image to the right). The observation time typically is several nights. The source of this radiation (gammas of > 100 GeV) is thought to be due to shock wave acceleration in the nebula. The gamma image can not compete, in resolution, with optical observations, but the existence of a VHE gamma flow, and its intensity, allow to develop models of both the source and the propagation of its emission through space.

The recording of images in a Cherenkov telescope has quite different constraints from what optical telescopes require. The phenomenon to be studied, the electromagnetic shower, is comparatively large, and the extraordinary resolution of CCDs (which have replaced photographic file in most optical telescopes) is not required, nor is the ultimate precision in the mirror surface. Instead, sensitivity to single photons and the best possible time resolution are important, because the signal is weak, and the discrimination against non-electromagnetic showers is helped by determining precise arrival times. Highest-quality photomultipliers are used, therefore, their size matched to the resolution of the shower. The pictures show the camera of one of the HEGRA telescopes, and (right) the light collecting surface in front of the camera, specifically developed to ensure optimal entrance angles for all incoming photons (so-called Winston cones).