Low light level detectors for astronomy.

There is an almost bewildering variety of detectors being used for ground-based astronomical observations. Many of the detectors have advantages for particular projects. One-dimensional detectors are simple, yet suitable for spectroscopy. Other detectors offer high photometric precision and dimensional stability. Some are designed for observing only faint objects and some for bright ones. Sometimes the necessity of having a high quantum efficiency at a particular wavelength dictates the choice of detector. In reality the situation is even more chaotic. This review has not covered some of the exotic detectors that have been developed for x-ray and far-ultraviolet imaging from spacecraft. These detectors use devices such as resistive plates, multianode or crossed-wire microchannel plates, and so on. These devices have not yet seen extensive use in ground-based applications, and their future as visible light detectors remains uncertain. The reasons for the development of such a wide variety of detectors are clear. Commercially available devices are simply not capable of meeting the low light level and photometric performance capabilities needed for astronomical observations. The driving forces are the commercial and military applications for detectors, and with few exceptions there has not been sufficient funding available to mount a detector development program for astronomy. Therefore, astronomers have sought to adapt existing commercial devices to the particular problem at hand. The large number of individual efforts summarized in this review is the result. In the future, I expect the variety to diminish as one or two really good detectors become capable of performing well under the wide variety of observational conditions encountered in astronomy. Many people have proclaimed the ultimate detector to be just around the corner. This is yet to happen. However, I venture to speculate that low-noise, high-performance CCD detectors with a format of 500 by 500 or larger will emerge as the preferred astronomical detector within the next few years. I also expect photographic plates will continue to be used for the many applications requiring wide fields up to 10,000 pixels on a side. Finally, as the detectors approach the ultimate quantum limit, attention will shift away from them and toward development of the system necessary to manipulate, display, and extract the information from the 250,000 numbers that make up a 500 by 500 digital image.