Stellar Colours
From the time of Hipparchus down to about 1860, all astronomy had been done with the eye as the detector. Starting about 1860, astronomy began using other detectors, first photography and then other kinds of devices. Today, most astronomical observations are done using solid state chips as detectors.
The eye has a particular response. It is most sensitive to light of a yellow-green colour, with lower sensitivity in the violet, blue and red. Photographic film, however, has a very different sensitivity. It is most sensitive in the blue-violet region. Without special treatment, a photograph is blind to green, yellow and red light. The signal that is produced by a detector is the product of two factors:
This product must be summed (or integrated) over the total range of wavelengths. Therefore, the eye and a photograph can produce very different signals for the same star.
The simplest case is for a white star, which is taken to have the same intensity at all wavelengths. Then, if the eye and the photograph sample the same amount of light, they will produce the same signal, even though they are sensitive to different spectral regions. A red star, however, radiates more strongly at longer wavelengths and more weakly at shorter wavelengths. Therefore, red star will appear brighter to the eye than to the photograph. The reverse is true for a blue star, which radiates more strongly at short wavelengths and more weakly at longer wavelengths. Therefore, we will get different signals, and hence different magnitudes, for the same star if we use different detectors. This situation is true for any detector that does not exactly match the sensitivity of the eye.
To address this situation, astronomers do two things. First, an observation is never done with a ``naked'' detector because detectors vary, particularly concerning the extent of their sensitivity. For example, different people have different abilities to see violet or red colours. To combat this problem, astronomers place standard filters in front of the their detectors. These filters are made with explicitly defined central wavelengths, peak transmission, and range of wavelengths transmitted.
With this change, our measurement of the light from an astronomical source is now the integral of the product of three functions:
Second, astronomers use the idea that white stars have constant brightness across all wavelengths. This idea is used to define the magnitude at every wavelength to be the same for white stars. In practice, this is done with a set of star we observe to be white. This turns out to be very close to what we would get if we use the bright star Vega and set all of its magnitudes to its visual (eye) magnitude.
Following these two steps, the magnitude of a particular star at, say, blue light is found by measuring the blue light from both the particular star and a white star. Because we have defined the white star's blue magnitude to be same as its visual magnitude, we have only one unknown, the blue magnitude of the star being studied.
Colour Index
This complicated procedure can be turned around to gain a quantitative measurement of stellar colour, something that is normally a subjective judgement. The colour index is defined to be the difference between the magnitudes at two different wavelengths. The convention is to subtract the longer wavelength magnitude from the short wavelength magnitude.
For a white star, the colour index is clearly 0.00. For a red star, however, the shorter wavelength magnitude is fainter than the longer wavelength magnitude. Keeping in mind that a fainter magnitude corresponds to a larger magnitude, the red star will have a colour index that is > 0.00. For a blue star the reverse is true. The shorter-wavelength magnitude is brighter or smaller than the longer-wavelength magnitude. As a result, the colour index of a blue star is < 0.00. The actual numerical value of the colour index tells us quantitatively how red or how blue the star is. For example, the bright star Rigel has a colour index of -0.03, meaning it is just slightly bluer than white. The sun has a colour index of +0.63, in keeping with its yellowish appearance. The bright red star Betelgeuse has a colour index of +1.85.
Standard Magnitudes and Colours
There are many different systems of magnitudes and colours in use in astronomy. Some of them are general purpose, while others are designed to detect specific things about the properties of the objects being studied. In most cases the systems were developed using photomultiplier detectors. These are vacuum tubes that contain a surface that is sensitive to light. A photon of light striking the surface within the vacuum tube releases an electron (called a photo-electron). The electron is accelerated by the voltage established inside the tube, and it is made to crash into sensitive surfaces inside the tube were it creates a shower of many electrons. In this way, each electron can be enhanced until it has produced as many as a million electrons by the time it reaches the base of the tube. The resulting current can be measured, which has the property of being directly proportional to the amount of light striking the tube initially.
Standard System of Johnson and Morgan - UBV System
The most widely used standard system was developed by Harold Johnson and W. W. Morgan in 1953. They selected three filters to use in front of their photomultiplier detectors.
| Filter | Central Wavelength | Wavelength Width |
| U (ultraviolet) |
366 nm | 70 nm |
| B (blue) |
440 nm | 92 nm |
| V (visual) |
553 nm | 85 nm |
The central wavelengths of the three filters were chosen to record certain information present in the light of stars. Because the bands are broad, they enable the measurement to be made to very faint magnitudes. However, the very width of the bands creates certain problems.
Standard System of Strömgren - uvby System
To address some of these problems, a second, general purpose standard system was developed. This was by Bengt Strömgren, about 1960. Strömgren spent some time at the Yerkes Observatory, where he had the opportunity to learn from W. W. Morgan about the problems that surfaced with the UBV system. He chose to use narrower filters and he also decided to use four of them.
| Filter | Central Wavelength | Wavelength Width |
| u (ultraviolet) |
350 nm | 30 nm |
| v (violet) |
411 nm | 19 nm |
| b (blue) |
467 nm | 18 nm |
| y (yellow) |
547 nm | 23 nm |
The narrower filters give much cleaner information about the radiation, but they restrict the application of the magnitudes to slightly brighter limits, although with the current sensitivity of detectors and the size of telescopes, this is not much of a limitation.
Note that the use of four filters makes it possible to define three colour indices: u-v, v-b, and b-y. In fact, Strömgren developed more complicated quantities by using two of the colour indices together. For example, he defined c1 = (u-v) - (v-b) and m1 = (v-b) - (b-y). These are designed to give information about some of the other aspects of the star's radiation.
The UBV and the uvby systems are just two of the many systems of filter magnitudes and colours. There is even one created at the University of Toronto known as the DDO system, named after our David Dunlap Observatory.