Instrumentation in astronomy
As in the case of all of the sciences, instrumentation has been developed in astronomy so that the data provided by the observations are no longer subjective. Again, as in other sciences, the application of instrumentation immediately revealed that the scope for measurement is also extended. For example, when Galileo employed the telescope for astronomical observation, a new range of planetary phenomena was discovered and the number of observable stars was greatly increased. Since Galileo’s time, the whole range of observable phenomena has continued to grow with the application of each new type of observing equipment.
The instrumentation which was first applied to astronomy was designed so that the actual measurement of record was made by eye. When photographic material became available, the range of possible observation was immediately increased. This has now been further extended by the introduction of solid state devices in the form of CCDs (charge coupled devices). Whereas the eye is capable of being able to concentrate on only a few stars at a time in a star field, the photographic plate or CCD chip is able to record the light from every star in the field simultaneously. For a star to be seen by eye, the brightness must be above a certain threshold: the eye is not able to accumulate the energy it receives over a period of time to form an impression. The photographic plate and CCD, however, are able to do this and, if a time exposure is made, the resultant images depend on the total energy which falls on to the detector. Thus, besides being able to record many images simultaneously, these devices allow faint stars to be recorded which would not normally be seen by eye (see figure1).
The variation of the sensitivity with wavelength of these detectors is also different to the eye. For example, photographic plates of different types are available with a range of spectral sensitivities. Some plates have their peak of sensitivity in the blue while others have their peak in the red. Bluesensitive plates will obviously give strong images for blue stars and not for the red, while red-sensitive plates give weak images for blue stars and strong images for red. By using two plates of different spectral sensitivity to photograph a star field, the fact that stars are coloured is easily demonstrated. Because of the physical process involved in the detection of radiation by a silicon-based solid state detector, the natural peak sensitivity tends to be in the red end of the spectrum but, again, the colour response of an applied detector can be modified at its manufacture.
Some special photographic materials are sensitive to colours which cannot be seen by the normal eye. The colour range of astronomical observations can be extended into the ultraviolet or the infrared by the choice of a particular photographic emulsion.
Thus, by recording the astronomical observation on a detector other than the eye, it is possible to extend the scope of the observation by looking at many objects simultaneously, by looking at a range of objects which are too faint for the eye to see and by looking at a much broader range of colour.
The range of available detectors has increased greatly since the photographic process was first applied to astronomy. Detectors based on the photoelectric effect have a common application. Detectors specially designed for infrared work can also be attached to optical telescopes. After the discovery that energy in the form of radio waves was arriving from outer space, special telescopes were designed with sensitive radio detectors at their foci and the era of radio astronomy was born. It is also apparent that our own atmosphere absorbs a large part of the energy arriving from outer space but, with the advent of high flying balloons and artificial satellites, these radiations are now available

for measurement. New branches of γ -ray, x-ray and infrared astronomy are currently increasing the information that we have concerning the extra-terrestrial bodies.
Although the large range of detectors removes to a great extent the subjectivity of any measurement, special care is needed to avoid the introduction of systematic errors. Each detector acts as a transducer, in that energy with given qualities falls on to the detector and is converted to another form; this new form is then measured. For example, when radiation falls on the sensitive area of a photocathode, the energy is converted in the release of electrons which can be measured as a flow of electric current. The strength of the incident energy can be read as the needle deflection on a meter or converted to a digital form for direct processing by a computer.
The process of converting the incident radiation to a form of energy which is more acceptable for measurement is never one hundred per cent efficient and it is essential that the observer knows exactly how the recording system responds to a given quality and quantity of radiation. In other words, the whole of the equipment which is used to make an observation must be calibrated. The calibration can be calculated either by considering and combining the effects of each of the component parts of the equipment or it can be determined by making observations of assumed known, well-behaved objects.  Because of the impossibility of having perfect calibration, systematic errors (hopefully very small) are likely to be introduced in astronomical measurements. It is one of the observer’s jobs to ensure that systematic errors are kept below specified limits, hopefully well below the random errors and noise associated with the particular experimental method.
Although every piece of observing equipment improves the process of measurement in some way, the very fact that the equipment and the radiation have interacted means that some of the information contained in the parameters describing the incident radiation does not show up in the final record and is lost. All the qualities present in the incident energy are not presented exactly in the record. Each piece of equipment may be thought of as having an instrumental profile. The instrumental profile of any equipment corresponds to the form of its output when it is presented with information which is considered to be perfect.
For example, when a telescope is directed to a point source (perfect information), the shape of the image which is produced (instrumental profile) does not correspond exactly with the source. The collected energy is not gathered to a point in the focal plane of the telescope but is spread out over a small area. The functional behaviour of the ‘blurring’ is normally referred to as the point spread function or PSF. For the best possible case, the PSF of the image of a point source is that of a diffraction pattern but inevitably there will be some small addition of aberrations caused by the defects of the optical system or blurring by atmospheric effects. If the recorded image is no larger than that of the instrumental profile, measurement of it gives only an upper limit to the size of the object. Detail within an extended object cannot be recorded with better resolution than the instrumental profile.
For any instrument, there is a limit to the ‘sharpness’ of the recorded information which can be gleaned from the incoming radiation. This limit set by the instrument, is frequently termed the resolving power of the instrument. In all cases there is an absolute limit to the resolving power of any given equipment and this can be predicted from theoretical considerations. Certain information may be present in the incoming radiation but unless an instrument is used with sufficient resolving power, this information will not be recorded and will be lost. When any given piece of equipment is used, it is usually the observer’s aim to keep the instrument in perfect adjustment so that its resolving power is as close as possible to the theoretical value.
As briefly mentioned earlier, as with all sciences involving quantitative observations, the measured signal carries noise with the consequence that the recording values are assigned uncertainties or errors. One of the ways of describing the quality of measurements is to estimate or to observe the noise on the signal and compare it with the strength of the signal. This comparison effectively determines the signal-to-noise ratio of the measurement. Values of this ratio may be close to unity when a signal is just about detectable but may be as high as 1000:1 when precision photometry is being undertaken.