Design Considerations

To achieve the greatest scientific return with VERITAS, the required characteristics of the chosen design for VERITAS are low energy operation ( $\mathrel{\hbox{\rlap{\hbox{\lower4pt\hbox{$\sim$ }}}\hbox{$<$ }}}$100GeV), large array collection area, and the ability to split the array into sub-arrays.

In general, the multiple stereoscopic view of the sky afforded by an array has several significant advantages over a single telescope. For example, the angular origin of the primary gamma-ray can be determined to better than about 0.1$^{\circ}$ which may be crucial for mapping extended sources such as SNR and eliminating source confusion when there are several possible candidates within the field of view. It dramatically reduces the background from isotropically distributed cosmic-ray protons and electrons. Also, the shower core location can be found to better than 10m which, when combined with the light in the shower striking several telescopes, allows one to reconstruct the energy with a factor of three better resolution than a single telescope. In principle, two telescopes are sufficient for stereoscopic reconstruction of a shower, but only a relatively small fraction of the showers are sufficiently near both telescopes to trigger them simultaneously, yet far enough away from the line joining the telescopes to permit good angular reconstruction. Thus, a three-telescope array is the minimum practical size. As more telescopes are added, the gain in sensitivity comes mostly through reducing the energy threshold at which meaningful measurements can be made. The joint requirements of adequate light from a single shower at the telescopes (implying a small separation) and angular reconstruction of the shower (implying a large separation) gives an optimal separation of about 80m for the telescope spacing. The collection area of an array grows more slowly beyond three telescopes and is related to the area of the shower (about 250m in diameter) folded with the area of the array.

The VERITAS configuration was chosen (after extensive simulations) for the reasons outlined in the remainder of this sub-section.